WO2008147859A2 - Self-assembled proteins and related methods and protein structures - Google Patents

Abstract

The present invention provides user-directed construction of novel specific homo- and hetero-dimeric, and multimeric assemblages of proteins. The present invention is comprised of gene sequences that transcribe peptide sequences that form links between proteins, where the peptide sequences produce a hook or loop which supports specific self-assembly of homo-dimers, hetero-dimers and multimers of the proteins to which they are attached. The hook or loop may have a short aliphatic repeat sequence and a metal binding loop. The present invention also provides a method of constructing a hook motif of metal binding loop sequences that may be attached to at least one aliphatic repeat sequence to produce the assemblages of proteins. Also provided are protein structures produced by the methods of the present invention.

Description

SELF-ASSEMBLED PROTEINS AND RELATED METHODS AND PROTEIN

STRUCTURES

[0001] The present application is a Continuation-in-Part application of United States Patent Application Number 1 1,751,850, filed May 22, 2007, entitled "Self- Assembled Proteins and Related Methods," which claims priority to United States Provisional Application Number 60/808,232, filed May 24, 2006, both of which are incorporated herein by reference.

BACKGROUND OF THE INVENTION

Field of the Invention

[0002] The present invention relates to protein self-assembly methods and, more particularly, to homo- and hetero-dimeric and multimeric assemblages of proteins and methods for their construction. Description of the Prior Art

[0003] Understanding protein-protein interactions is critical to the understanding of how proteins participate in biological regulatory networks. All cellular signaling pathways rely on the joining or breaking of protein interactions to maintain correct function (Fry, D.C., Biopolymers, Peptide Science, 84:535-552, 2006) Protein- interactions have been defined to have different quaternary structures in which the simplest model is dimerization. In the search for new active agents, pharmaceutical companies are targeting therapeutic agents towards the interface of dimerization complexes (Tropsha, A.B. et al., PNAS USA, 88:9488-9492, 1991). In most cases, the goal is to disrupt protein-protein interactions at specific points within a biosignaling pathway (Graddis, T.J.M. et al., Biochemistry, 32:12664-12671, 1993). While knowledge on the subject is substantial and the pharmaceutical industry continues to develop drugs for inhibiting protein-interactions, there still is much to learn about how organisms organize the hundreds of interactions that take place at one time.

[0004] Dimerization is a critical structural feature of a myriad of proteins and the regulatory processes in which they participate. While the function of some proteins requires that they be in a dimeric form, the formation may not be directly involved in regulation. Dimerization is, however, involved in signal transduction pathways, repressor and transcription factor operations, enzyme activation and cell-to-cell communication. Dimer formation may occur prior to the signaling or regulatory event or the dimerization itself may be an integral part of the regulatory process. [0005] At physiological concentrations, many dimers are in equilibrium with their monomer components. The dimers are bound together through an interface stabilized by a mixture of hydrophobic interactions, charge attractions and hydrogen bonds. Manipulation of the monomer-dimer equilibrium through ligand binding is a normal part of metabolism. Conformational changes induced by the binding of a ligand is a common way of promoting dimerization and, thereby, regulating the function of the proteins. The complexity of ligands ranges from simple ions to small signal molecules to complex proteins. Metalloregulation, the regulation of processes by the complexation of metal ions to proteins, is common in metal metabolic pathways and some oxidative stress responses. Generally, in metal loregulatory events the binding of metal ions to proteins is thought to induce or stabilize a conformational change in the structure to regulate its activity.

[0006] Conventional technology for creating dimers or multimers from proteins that normally are not coupled requires that the proteins be (1) chemically cross-linked, (2) created as direct gene fusions of the proteins involved or (3) indirectly linked by creating gene fusions with specific protein-protein binding motifs so that they, when translated, would be expected to specifically bind to each other through the added motifs with or without the aid of an exogenous regulatory ligand. [0007] Generally, chemical cross-linking requires that the proteins be isolated and either (1) mixed together and treated with reagents to cause them to be covalently attached to each other or (2) each isolated protein be modified with different reagents that will enable the modified proteins to interact when mixed. In the former case, it is difficult to specifically dimerize the proteins into a single or limited series of structures because the crosslinks can form at multiple locations on the surfaces of the two proteins. In the latter case, each protein is modified separately with compounds that will couple with the other when mixed and thereby link the two proteins. Neither of these methods produces entirely specific quaternary structural links between the proteins. Both can lead to conformational changes in the proteins being modified and thus perturb their normal function and in some cases cause the complexes to be poorly soluble. Chemically cross-linked proteins are, however, commonly used in in vivo applications by first isolating the proteins, chemically cross-linking them and then injecting them into an organism. Chemical cross-links frequently are used to link enzymes to antibodies, which subsequently are used analytically in ELISA, tissue fixing or other in vitro analyses.

[0008] Chemical cross-linkers capable of diffusing across cell membranes have been used to study the state of oligomerization of dimeric and hexameric species and to probe for heterocomplexes. These studies are best performed on well characterized proteins so that appropriate linkers can be used. The weakness of these methods is that it is not possible to entirely limit the modification to only those proteins being targeted.

[0009] Though there are in vitro cleavable chemical cross-linkers, equilibrium between the monomer and dimer cannot be maintained, manipulated or used as a switch. Few of the modifying species can serve as a tool for subsequent affinity purification. Biotinylation, which has been an important tool for in vitro affinity purification (Pierce), cannot be used to purposely construct dimers or higher order species in vivo.

[0010] Pierce Biotechnology Inc. markets a wide range of chemical cross-linkers and bioconjugate linkers that interact with a variety of functional groups such as amino, sulphydryls and carboxylates. Some of these cross-linkers diffuse across membranes, but their reactions are largely non-specific and target a broad number of proteins within a cell.

[0011] Directly fusing two genes so that they are expressed as one fusion protein is commonly used to add an affinity tag to a protein. The affinity tag then enables the protein fusion to be purified more easily or, in some cases, used as an analytical tool to detect or measure the protein. Proteins, such as glutathione transferase (Pharmacia) and maltose binding protein (NEB); protein domains, such as S-tag (Novagen) and FLAG (Kodak); and 6 histidine repeats (His tag) frequently are used for affinity purification purposes. Genes or gene fragments also are used to target the fusion protein to specific cellular locations. Novagen (and others) markets a vector that produces a gene fusion of the protein of interest to an export sequence so that the nascent protein will be excreted into the periplasmic space.

[0012] The S-tag, FLAG and, especially, the His tag are not as prone to cause folding problems in their fusion partners. However, it is relatively common for the resultant large glutathione transferase and maltose binding protein fusions to have solubility and expression problems. The His tag, which is the most commonly used affinity purification tag on the market today, it the most advantageous affinity tag because it works through a coordination complex and can be used in denaturing conditions. However, although very useful for affinity methods, all of these methods do not enable the construction of homodimers or specific heterospecies. [0013] The direct fusion of the domains derived from the FKDP protein to the proteins of interest can be used to produce homodimers (ARIAD Pharmaceuticals, Inc.) under the control of exogenous rapamycin or related compounds due to the FKDP protein's high affinity for rapamycin. Two FKDP domains bind to each molecule (ligand) or rapamycin. Similarly, using hetero-fusions to FKDP and FRB, domain- specific heterodimers can be induced to form through the addition of a modified rapamycin (ARIAD Pharmaceuticals, Inc.). The regulated feature of these systems is useful and has been used in trafficking studies and may find uses in drug delivery systems. The size and complexity of the fusion proteins, however, pose problems in some cases. The delivery of rapamycin to the cell, tissue or organisms can be difficult. As designed, the monomer to dimer affinity cannot be modified. [0014] There are several cloning systems, so called two-hybrid systems, where the possibility of protein-protein interactions between heteromers can be probed through the formation of dimeric species (Invitrogen and others). The two-hybrid systems also can be used to explore the interactions between specific dimers. [0015] While these systems excel at probing for interactions between one protein and a library of others, they do not enable the formation of homo-dimers of a protein or hetero-dimer formation between proteins selected by the user. The system is, by its nature, linked to a reporter system. In addition, the two-hybrid systems cannot be used to build higher order species, establish an equilibrium between the monomer and dimer, etc., or be regulated and used in trafficking studies.

[0016] The fusion of genes to a gene or gene fragment, as proposed for the hook motif, enables the formation of homo-dimers in vivo. The possibility of producing fusions of this type has been explored through manipulation of classical leucine- zippers and zinc-fingers. Both of these motifs require a larger sequence be attached to the desired proteins. Both the zipper and finger motifs cannot have their monomer- dimer equilibrium easily manipulated, be used as a switch, be used readily in affinity purification strategies or be used easily to make discrete multimer complexes. To the best of the inventor's knowledge, neither the leucine-zipper nor the zinc finger motifs has been utilized in commercial applications, although the leucine-zipper motif has been explored by an industry group.

[0017] With respect to the direct purification of fusion proteins by affinity methods, there are many choices. His-tag (Pharmacia, Sigma and others), S-tag (Novagen), FLAG (Kodak), GST-tag (Pharmacia and others) and maltose binding protein tags all are used in gene fusion based systems to aid in the purification of bacterial and yeast over-expression systems, but these systems are not used in vivo or in vitro to construct dimer or higher order structures and for the most part they are not suited to that task.

[0018] There exists a need, therefore, for sequence ("hook") motifs useful as linkers as well as for a method to link proteins to form specific hetero- and homo- dimeric and multimeric protein structures in the living body in vivo and in a test tube or apparatus in vitro.

SUMMARY OF THE INVENTION

[0019] It is an object of the present invention to provide gene sequences for forming homo-dimers, hetero-dimers and multimers of proteins in vivo and in vitro. [0020] It is another object of this invention to provide a method of linking proteins without conformational changes in the proteins.

[0021] It is a further object of this invention to provide a method of linking proteins that enables the construction of homodimers of proteins or construction of heterodimers between proteins.

[0022] It is another object of the present invention to provide protein structures produced by the methods of the present invention.

[0023] The above needs are met and objects accomplished by providing a peptide sequence, said peptide sequence comprising at least one component comprised of a CxCx(₄.6)CxC (SEQ ID NO: l)-metal binding loop, wherein the peptide sequence produces peptides or proteins. The CxCx^CxC (SEQ ID NO: l)-metal binding loop can be attached to aliphatic repeat groups, which increases the stability of the component. An example of a CxCxø^CxC-metal binding loop is, without limitation, CxCxxxxCxC (SEQ ID NO. 1). The peptides or proteins produced therefrom form homo-dimers, hetero-dimers and multimers of proteins both in vivo and in vitro. The peptide sequence comprising at least one component comprised of a CxCx^_-6)CxC (SEQ ID NO: l)-metal binding loop and aliphatic repeat groups can include, for example, the following sequence:

ATLTOEDIQOIMKOLNKKEPVETIECNCIPGQCECKKO (SEQ ID NO: 2) [0024] The present invention also provides a genetic sequence that transcribes the peptide sequence comprised of the CxCx^_-6)CxC (SEQ ID NO: l)-metal binding loop and aliphatic repeat groups.

[0025] The present invention further provides a method of constructing assemblages of proteins with linking between the proteins. The method is comprised of forming a hydrophilic sequence ("hook") motif, comprised of a metal binding loop sequence attached to at least one aliphatic repeat sequence; providing a plurality of proteins; and mixing the plurality of proteins so that they self-assemble in the presence of the metal binding loop in order to form protein structures, such as, without limitation, homo-dimers, hetero-dimers or multimers.

[0026] The attachment of the metal binding loop sequence to the at least one aliphatic repeat sequence allows for the formation of specific interactions of the plurality of proteins. The aliphatic repeat sequence contains residues which contribute to the stability and specific interaction of the plurality of proteins. [0027] The metal of the metal binding loop is, for example and without limitation, zinc [Zn(II)] or copper [Cu(I)].

[0028] The homo-dimers, hetero-dimers and multimers are effective for both in vivo and in vitro uses.

[0029] The present invention still further provides protein structures produced by the methods of the present invention, in which the protein structures are characterized as being suitable for use for, without limitation, diagnostic reagents; biomarkers; metal-activated switches; cell-trafficking studies; affinity purification of in vivo constructs; nanoscale construction; for imaging of cells and tissues such as, without limitation, visible imaging, fluorescent imaging and confocal imaging; delivery of proteins or pharmaceuticals having said proteins bound thereto to specific tissues or cells; purifying and separating compounds; cell research; and discovery of compounds for treating disease. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] The Patent or Application File contains at least one drawing executed in color. Copies of this patent or patent application publication with color drawing(s) will be provided by the office upon request and payment of the necessary fee. [0031] A full understanding of the invention can be gained from the following description of the preferred embodiments when read in conjunction with the accompanying drawings in which:

[0032] Figure 1 shows the hook motif according to e embodiments of the present invention;

[0033] Figure 2 shows realized and potential hook motifs according to embodiments of the present invention;

[0034] Figure 3 shows the hook motif according to embodiments of the present invention, in which (A) contains a CxCxxxxCxC (SEQ ID NO: 1) motif and an aliphatic repeat; (B) shows fusion proteins constructed to express the motif; and (C) shows proteins expressing the motif and connecting to make dimers; [0035] Figure 4 shows hook motifs attached to solid resins for capturing proteins expressing the motifs according to embodiments of the present invention; and [0036] Figure 5 shows motifs with varying affinities, in which (A) shows a hetero- trimer constructed with fusions containing four distinct hook motifs; and (B) shows a metal binding loop with an attached N-terminal fluorescent tag used as a probe; [0037] Figure 6 shows a pWH6 construct for the expression of 6x his CopY; [0038] Figure 7 shows a pWY145 construct for the expression of 6xhis CopY; [0039] Figure 8 shows a sequence of a synthetic gene, as well as the translated fusion protein sequence, according to embodiments of the present invention; [0040] Figure 9 shows a plasmid construct for the expression of 6xhis GBl- Ymbs38;

[0041] Figure 10 shows a first molecular weight standard curve for a Shodex KW 803 HPLC column;

[0042] Figure 11 shows a second molecular weight standard curve for a Shodex KW 803 HPLC column;

[0043] Figure 12 shows an excel formula spreadsheet for large zone chromatography;

[0044] Figure 13 shows large zone calculations for Zn(II)GB 1-Ymbs38; [0045] Figure 14 shows gel filtration chromatography of GB1-Ymbs38;

[0046] Figure 15 shows ESI mass spectrum of Zn(II)GB 1-Ymbs38;

[0047] Figure 16 shows a method for calculating ion charge and molecular mass from the mass spectrum;

[0048] Figure 17 shows ESI mass spectrum of GBl;

[0049] Figure 18 shows ESI mass spectrum of apo-GB 1 -Ymbs38;

[0050] Figure 19 shows thrombin cleavage of GB1-Ymbs38;

[0051] Figure 20 shows a schematic of HIS-Select Resin Binding Assay. (A) shows a 6x histidinte tagged (6xhis) version of the protein mixed with an untagged protein; (B) shows the protein mixture applied to the HIS-Select affinity resin; and

(C) shows imidazole added to elute the histidine-tagged protein from the resin;

[0052] Figure 21 shows SDS-PAGE of HIS-Select Resin Binding Assay with GBl-

Ymbs38 Variants;

[0053] Figure 22 shows large zone size exclusion chromatography on Zn(II)CopY.

(A) is a chromatograph showing the 280 nm absorbance trace for the large zone experiment; and (B) shows the first derivative cureves of the elution profiles from

(A);

[0054] Figure 23 shows a comparison of large zone chromatography assays on

GB1-Ymbs38;

[0055] Figure 24 shows non-linear least squares best fit of large zone chromatography data;

[0056] Figure 25 shows a residual plot of large zone chromatography data for apo-

GB1-Ymbs38; and

[0057] Figure 26 shows analytical ultracentrifugation data to confirm size exclusion data. (A) shows a 85%- 15% Zn(II)CopY dimer-monomer mixture and (B) shows a

25%-75% an apoCopY dimer-monomer mixture without Zn(II);

DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0058] The present invention provides homo- and -hetero-dimeric and multimeric self-assembly of proteins and methods for their construction.

[0059] In an embodiment of the present invention, there is provided a peptide sequence, said peptide sequence comprising at least one component comprised of a CxCx(_4-6)CxC (SEQ ID NO. l)-metal binding loop, wherein the peptide sequence produces peptides or proteins. The CxCx^CxC (SEQ ID NO: l)-metal binding loop can be attached to aliphatic repeat groups, which increases the stability of the component. An example of a CxCx^_-6)CxC (SEQ ID NO. l)-metal binding loop is, without limitation, CxCxxxxCxC (SEQ ID NO: 1). The peptides or proteins produced therefrom form homo-dimers, hetero-dimers and multimers of proteins both in vivo and in vitro. The peptide sequence comprising at least one component comprised of a CxCx(₄-₆)CxC (SEQ ID NO: l)-metal binding loop and aliphatic repeat groups can include, for example, the following sequence:

ATLTQEDIQQIMKQLNKKEPVETIECNCIPGOCECKKO (SEO ID NO: 2) [0060] In another embodiment of the present invention, there is provided a genetic sequence that transcribes the peptide sequence comprised of the CxCx_(4-6)CxC (SEQ ID NO: l)-metal binding loop and aliphatic repeat groups.

[0061] In a further embodiment of the present invention, there is provided a method of constructing assemblages of proteins with linking between the proteins. The method is comprised of forming a hydrophilic sequence ("hook") motif, comprised of a CxCx(₄-₆)CxC (SEQ ID NO: l)-metal binding loop sequence, which can attach to at least one aliphatic repeat sequence; providing a plurality of proteins; and mixing the plurality of proteins so that they self-assemble in the presence of the metal binding loop in order to form protein structures, such as, without limitation, homo-dimers, hetero-dimers or multimers.

[0062] A non-limiting example of the CxCx_(4-6)CxC (SEQ ID NO: l)-metal binding loop sequence is CxCxxxxCxC (SEQ ID NO: 1).

[0063] The attachment of the metal binding loop sequence to the at least one aliphatic repeat sequence allows for the formation of specific interactions of the plurality of proteins. The aliphatic repeat sequence contains residues which contribute to the stability and specific interaction of the plurality of proteins. [0064] The metal of the metal binding loop is, for example and without limitation, zinc [Zn(II)] or copper [Cu(I)].

[0065] The homo-dimers, hetero-dimers and multimers are effective for both in vivo and in vitro uses.

[0066] In still a further embodiment of the present invention, there is provided protein structures produced by the methods of the present invention, in which the protein structures are characterized as being suitable for use for, without limitation, diagnostic reagents; biomarkers; metal-activated switches; cell-trafficking studies; affinity purification of in vivo constructs; nanoscale construction; for imaging of cells and tissues such as, without limitation, visible imaging, fluorescent imaging and confocal imaging; delivery of proteins or pharmaceuticals having said proteins bound thereto to specific tissues or cells; purifying and separating compounds; cell research; and discovery of compounds for treating disease.

[0067] Suitable aliphatic repeat sequences may be, without limitation, hydrophobic amino acids such as valine, isoleucine, proline, alanine, methionine, tyrosine, phenylalanine or synthetic residues having similar hydrophobic properties. The aliphatic repeat sequences may repeat in the peptide sequence, for example and without limitation, about every four amino acid residues.

[0068] Hook motifs are shown in Figure 2. Residues around cysteines are responsible for the specificity of the interactions. The top sequence in Figure 2 has been fully tested and the highlighted portion shows hydrophobic residues, frequently as aliphatic repeats. The cysteines in the metal binding motif are in bold. Four of the genes containing these sequences have been cloned.

[0069] The aliphatic repeats support dimerization even in the absence of the metal binding loop, constituting the bulk of the stabilization. However, in the absence of a metal binding loop, there is a higher probability of forming non-specific interactions and non-specific hetero-dimers and multimers. The residues around the aliphatic repeats also contribute to the specificity. Both parts of the peptide sequence are involved in the specificity and stabilization of the protein-protein interactions. To construct dimers or multimers, the DNA sequence that codes for the peptides must be fused to the gene or genes of interest at one of the two termini or, possibly, between two domains. Homo-dimers are formed by fusing a single DNA sequence coding for a peptide motif to the gene of interest. The resultant protein will self assemble into a specific dimer in the presence of zinc(II) or copper(I). Formation of specific hetero species requires that two different motifs, which have a higher affinity for each other than for themselves, be used with the genes of interest. In the absence of metal ions, the motif is prone to more non-specific interactions with similar aliphatic motifs to make dimers or multimers. Most cells maintain an intracellular zinc concentration more than sufficient to insure that zinc is bound to the sequence. Figure 2 shows only a partial selection of the sequences that can be used for these operations according to the embodiments of the present invention. [0070] The advantageous features provided by the metal binding peptides (i.e., the "sequence motif or "hook motif) of the present invention are: (1) the sequences are short, and thus they are less likely to interfere with the folding of their fusion partners; (2) the sequences are hydrophilic, and thus they aid, not decrease, the solubility of the fusions; (3) the sequences can be used to form dimers and/or multimers for in vivo and in vitro applications; (4) it is possible to select a range of dissociation constants by varying the length of the aliphatic repeat region; (5) it is possible to build a metal- activated dimer (i.e., metal ions are easily administered); (6) due to the variety of sequences available, a large number of distinct constructs can be produced; (7) the sequences can be used as an affinity tag with similarly tagged proteins, in which the tag then can be used to purify or detect fusion proteins.

[0071] The most advantageous feature of the hook motif is the ability to use it to form specific user-directed self-assembling hetero-dimers in vivo. In natural settings, the motif aids in or supports dimerization. The hook motif is found at the termini of proteins and in separate regions or domains of proteins and can, in principle, be utilized in similar positions when used to make novel dimers.

[0072] Specific dimerization or the formation of higher order structures is difficult in that it requires the use of large scale structures to interface with one another. One benefit of the sequence motif of the present invention is that it is small and hydrophilic and, therefore, is less likely to interfere with the structural fold and function of the proteins being dimerized. In this case, the dimerization motif is small and the interaction is weakened by the removal of the metal ion. The opportunity to use this feature in living cells or in the construction of nanoscale fabrications outside of cells is unique. The potential to use the metal-binding-dimerization motif in the direct purification of the fusion proteins also is a significant benefit.

[0073] Various ways that the hook motif of the present invention can be employed are shown in Figures 3, 4 and 5.

[0074] In Figure 3A, the hook motif is shown to contain a CxCxxxxCxC (SEQ ID NO: 1) motif and an aliphatic repeat. Figure 3B shows fusion proteins constructed to express the motif. Figure 3C shows proteins expressing the motif and connecting to make dimers. Any proteins expressing the hook motif can participate in the dimer interaction. [0075] Hook motifs attached to solid resins can be utilized to capture proteins expressing the hook motifs. For example, in Figure 4A, hooks are attached to a solid matrix or contrast agent to form an affinity matrix for fusion proteins containing the hook tag attached. In Figure 4B, fusion proteins are captured by a solid matrix with hooks attached.

[0076] Figure 5A shows a hetero-trimer fused protein constructed with fusions containing four distinct hook motifs. Thus, the construction of motifs with varying affinities enables specific higher order structures to be built. Figure 5B shows a metal binding loop with an N-terminal fluorescent tag attached so that it can be used as a probe. In this way, dye molecules attached to a motif can bind to other molecules that contain the hook motif for labeling, cellular trafficking or delivery studies. [0077] There are 48,218 known peptide sequences in the human genome. Of these peptide sequences, 175 have the sequence motif of the present invention. Many of these peptide sequences are different splice variants of the same gene. Thus, there are approximately eighty-five unique human genes that have the sequence motif of the present invention. A blast analysis demonstrates that slightly less than half of the genes, i.e., thirty-seven genes, do not share significant homology to other genes with the sequence motif. The gene families contain between two and nine genes. The biggest family is the metallothioneins, followed by the heparan-sulfate 6-0 sulfotransferases. In the mouse genome, there are eighty-six unique genes with the sequence motif of the present invention. Table 1 provides a non-limiting example of eighty-five human genes which may contain the sequence motif for the CxQ_4-6)CxC (SEQ ID NO: l)-metal binding loop and aliphatic repeat groups of the present invention, including their names and description.

Table 1. Human Genes Containing Sequence Motif for CxCxxxxCxC-Metal Binding Loop and Aliphatic Repeat Groups

EnsemblPeptide_ID ID Gene Description

Chloride channel protein 6

ENSP00000234487 CLCN6JHUMAN CLCN6 (CIC-6)xxxxx

CDNA FLJ38312 fis,done ENSP00000354348 Q8N962JHUMAN FCBBF3021506

Novel protein, lipase A2 ENSP00000247992 Q5R387JHUMAN family predicted protein, similar to

HS6ST1 (heparan-sulfate 6-

ENSP00000344207 728969 O-sulfotransf erase)

ENSP00000363889 Q5T3S7JHUMAN RAP1 GTPase activating protein 1 ENSP00000354614 Q9UI23_HUMAN

Leptin receptor precursor (LEP-R) (OB receptor) (OB-

ENSP00000330393 LEPRJHU MAN LEPR R) (HuB219) (CD295 ENSP00000354769 Q5T750JHUMAN Late envelope protein 7 (LEP7).

Protein FAM5B precursor (BMP/retinoic acid-inducible

ENSP00000355364 FAM5B_HUMAN FAM 5 B neural-specific Protein FAM5C precursor ENSP00000294697 FAM5C_HUMAN FAM 5C (DBCCRl-like protein 1). Synaptotagmin-2 ENSP00000356236 SYT2_HUMAN SYT2 (Synaptotagmin II) (SytII) DnaJ homolog subfamily C member 5G (Gamma

ENSP00000296097 DNJ5G_HUMAN DNAJC5G cysteine string protein) CDNA PSEC0134 fis, clone

ENSP00000354988 O8NBL2 HUMAN PLACE1004757 hypothetical protein ENSP00000355121 NP 848590.3 LOCI 50771 isoform 1 Heparan-sulfate 6-0- sulfotransferase 1 (EC

ENSP00000259241 H6ST1_HUMAN HS6ST1 2.8.2.-) (HS6ST-1) solute carrier family 23

(nucleobase transporters),

ENSP00000295738 NP_653313.2 member 3

Collagen alpha-3(IV) chain precursor (Goodpasture ENSP00000327594 CO4A3 HUMAN COL4A3 antigen)

Amphoterin-induced protein 3 precursor (AMIGO-3) ENSP00000323096 AMGO3JHUMAN AMIGO3 (Alivin-3)

Immunoglobulin-like domain- ENSP00000345667 Q86SU0-3 ILDRl containing receptor 1 precursor cysteine-rich PAK1 inhibitor, ENSP00000372402 Q6ZQR7JHUMAN CRIPAK CDNA FLJ45978fιs, clone PROST2007444

Mutant fibroblast growth factor

ENSP00000323596 Q9NRB6JHUMAN FGFR3 receptor 3

Hypothetical protein

ENSP00000372328 Q5GMH4JHUMAN DKFZp566H184

ENSP00000320793 441019 ENSP00000330525 441019

Tenβurin-3 (Ten-3) (Tenasάn-M3) (Ten-m3) (Protein Odd Oz/ ten -m

ENSP00000373695 O9P273 HUMAN ODZ3 homolog 3). Cadherin-18 precursor ENSP00000274170 CAD18 HUMAN CDH18 (Cadherin-14) NEDD4 family-interacting protein 1 (Breast cancer-

ENSP00000322211 Q9BT67-2 NDFIPl associated protein SGA-1M).

Teneurin-2 (Ten-2)

(Tenascin-M2) (Ten-m2)

(Protein Odd Oz/ten-m

ENSP00000373555 Q9NT68JHUMAN ODZ2 homolog 2) fibroblast growth factor

ENSP00000366412 O59F30 HUMAN FGFR4 receptor 4

Laminin subunit alpha-2 precursor (Laminin M chain)

ENSP00000347393 LAMA2_HUMAN LAMA2 (Merosin heavy

F-box only protein 5 (Early

ENSP00000356208 FBX5JHUMAN FBXO5 mitotic inhibitor 1)

Regulator of G-protein

ENSP00000356194 RGS17 HUMAN RGS17 signaling 17 (RGSlZ)

ENSP00000369936 Q9UMK5JHUMAN ELN Elastin

Homo sapiens laminin, beta

ENSP00000205386 NM 007356.1 4 (LAMB4), mRNA

CDNA FU45737fis, clone JCMLC2002751, weakly similar

ENSP00000353120 Q6ZS87JHUMAN to Von Willθbrand factor SCO-spondin homolog (Bos

ENSP00000262089 Q76B61JHUMAN SSPO taurus), protein coding regulator of G-protein signalling

ENSP00000344630 O76081-2 RGS20 20 ENSP00000373980 NP_001008495.1 transmembrane protein 64 ENSP00000348109 Q71RG6JHUMAN

Protein FAM5A precursor (Deleted in bladder cancer

ENSP00000265922 FAM5A HUMAN FAM5A 1 protein)

Deleted in bladder cancer protein 1 precursor - duplicate

ENSP00000363075 O60477-2 DBCl ofFAM5A above Hypothetical protein

ENSP00000351967 NP 945352.1 LOC375791

Glutamate decarboxylase 2 (EC 4.1.1.15) (Glutamate

ENSP00000259271 DCE2 HUMAN GAD2 decarboxylase 65 Ret protein precursor, ret proto-oncogene (multiple endocrine neoplasia and medullary thyroid carcinoma 1, Hirschsprung

ENSP00000347942 Q9UQV8_HUMAN RET disease) ENSP00000371696 Q2TUQ5JHUMAN Mucin-6 precursor Mucin-2 precursor

ENSP00000351956 MUC2 HUMAN MUC2 (Intestinal mucin-2) Mucin-5B precursor (Mucin- 5 subtype B,

ENSP00000343037 MUC5BJHUMAN MUC5B tracheobronchial) (High CDNA FU46346 fis, clone TESTI4047328, moderately similar to M us musculus

ENSP00000341666 Q6ZRI0JHUMAN 070G otogelin

Keratin-associated protein 5-8 (Keratin-associated

ENSP00000310492 KRA58JHUMAN protein 5.8) ENSP00000374650 O9P2P4 HUMAN Kl AA1302 protein nuclear receptor co¬

ENSP00000254404 01542 IJHUMAN NCOR2 mpressor 2 Heparan-sulfate 6-0- sulfotransferase 3 (EC

ENSP00000330895 H6ST3_HUMAN HS6ST3 2.8.2.-) (HS6ST-3) CDNA FU43399 fιs, done ENSP00000370085 Q6ZUR6_HUMAN OCBBF2009926, HERV-K_16p3.3 provirus ancestral Env polyprotein

ENSP00000346747 ENK14_HUMAN (Envelope

Metallothionein-4 (MT-4) (Metallothionein-IV) (MT- ENSP00000219162 MT4_HUMAN MT4 IV)

Metallothionein-1E (MT-IE) (Metallothionein-IE) (MT- ENSP00000307706 MT1E_HUMAN MTlE IE)

Metallothionein-1M (MT-IM) (Metallothionein-IM) (MT- ENSP00000369146 MTlM HUMAN MTlM IM)

Metallothionein-1A (MT-IA) (Metallothionein-IA) (MT-

ENSP00000290705 MTlAJH U MAN MTlA IA) ENSP00000262499 NP_783319.1 Metallothionein M similar to metallothionein

ENSP00000369144 441771 IG

Metallothionein-1F (MT-IF) ENSP00000334872 MT1FJHUMAN MTlF (Metallothionein-IF) (MT-IF)

Metallothionein-1G (MT-IG) ENSP00000369139 MT1GJHUMAN MTlG (Metallothionein-IG) (MT- IG)

CDNA FUl 2986 as, clone

ENSP00000367588 XR_016086.1 NT2RP3000055

CDNA FLJ 12547 fis, clone ENSP00000330035 Q9H9U3JHUMAN NT2RM4000634. solute carrier family 5

(sodium/glucose cotransporter), member 10

ENSP00000284168 NP_689564.3 isoform 1

CDNA FU36000 fis, clone ENSP00000354266 Q8NA00_HUMAN TEST/2015180

Mitochondrial Rho GTPase

1, (Ras homolog gene family member Tl) (Ra c-

GTP-binding protein -like

ENSP00000351132 Q8IXI2-3 RHOTl protein).

Mitochondrial Rho GTPase 1

(MIRO-I) (hMiro-1) (Ras homolog gene family member Tl) (Rac-GTP-

ENSP00000346215 Q8IXI2-4 RHOTl binding protein -like protein

Laminin subunit alpha-1 ENSP00000374311 LAMA1_HUMAN LAMAl precursor (Laminin A chain)

CDNA FU41036fιs_l done ENSP00000371515 Q6ZWI3_HUMAN HLUNG2003872

Ran -binding protein 3

ENSP00000315894 O9H6Z4-4 RANBP3 (RanBP3) ENSP00000318233 Q8WYX9_HUMAN STXBP2 syntaxin binding protein 2

Lipolysis-stimulated

ENSP00000262627 Q86X29-2 LSR lipoprotein receptor

CDNA FLJ44760 fιs, done ENSP00000366491 Q6ZTD6_HUMAN BRACE3031579

Agouti signaling protein precursor (ASP) (Agouti

ENSP00000217422 ASIP_HUMAN ASIP switch protein)

Putative metallothionein ENSP00000246198 CT127_HUMAN C20orfl27 C20orfl27

Uncharacterized protein ENSP00000362061 CT111_HUMAN C20orflll C20orflll

Probable phospholipid- transporting ATPase HA (EC

ENSP00000342481 ATP9A_HUMAN ATP9A 3.6.3.1) (ATPase

DnaJ homolog subfamily C member 5 (Cysteine string ENSP00000358930 DNJC5_HUMAN DNAJC5 protein) (CSP)

Regulator of G-protein signaling 19 (RGS19) (G-

ENSP00000348338 RGS19 HUMAN RGS 19 alpha-interacting ENSP00000369962 Q9NSI5_HUMAN IGSFδprotein

CDNA FLJ11556 fis, clone

ENSP00000349329 Q9HAJ0_HUMAN HEMBA1003079

Chloride channel protein 5

ENSP00000343213 CLCN5_HUMAN CLCN5 (CIC-5)

ENSP00000354667 NP_055068.2 odz, odd Oz/ten-m homolog 1 ENSP00000324617 H6ST2_HUMAN HS6ST2 Heparan-sulfate 6-0- sulfotransferase 2 (E?C 2.8.2.-) (HS6ST-2)

[0078] There are just over one million known and predicted peptide sequences in the prokaryotic genome database (complete genome sequences only), which contains 352 species. Of these peptide sequences, only 158 contain the sequence motif of the present invention. Therefore, less than 0.02% of prokaryote peptides have the sequence motif, whereas almost 0.4% of human peptides contain the sequence motif. The Cys-x-Cys-xxxx-Cys-x-Cys (SEQ ID NO: 1) sequence motif appears, therefore, to be prevalent throughout the biological spectrum, being more common in mammals than in prokaryotes.

[0079] Table 2 shows the abundance of CxCx(_4-6)CxC (SEQ ID NO: l)-containing peptides from complete genome sequence databases.

Table 2. Abundance of CxCx_(4-6)CxC (SEQ ID NO: l)-containing peptides from complete genome sequence databases

(a) Known and predicted peptide sequences. Includes multiple peptides derived from the same gene, and likely contains false-positive gene predictions.

(b) Includes mostly unique genes. An attempt was made to remove duplicates and multiple peptides from the same gene. [0080] Table 3 shows examples of protein families with the CxCx(_4-6)CxC motif.

Table 3. Examples of protein families with the CxCx^CxC (SEQ ID NO: 1) motif

[0081] The consensus sequences for each of the protein families were created through cross-species alignments followed by analysis using programs such as MEME and WebLogo. The motif is found across all three domains of life in a wide variety of proteins. The spacing between the second and third cysteines varies, but often contains a proline, suggesting a loop structure.

[0082] Table 4 shows some an example of the evolutionary history of CxCx^- ₆₎CxC- (SEQ ID NO: 1) containing proteins. The presence of the CxCx^₅₆₎CxC (SEQ ID NO: 1) motif is restricted to a few phylogenic lineages of the widespread S- adenosyl-methyl-transferase (SAM) protein. Table 4. Example of evolutionary history of CxCx^CxC-containing proteins.

ATCC 25586 700970

j jeejjuunnii ssuubDsspp.. j je< juni NCTC 11168 33889

str. Nichols

[0083] Table 5 shows another example of the evolutionary history of CxCx^CxC (SEQ ID NO: l)-containing proteins. The CxCx₍₄,₆)CxC (SEQ ID NO: 1) motif is conserved in the Eukaryotic Heparan-sulfate-6-sulfotransferase family from humans to worms.

Table 5. Another example of evolutionary history of CxCx^CxC^SEQ ID NO: 1) containing proteins

62 • Homo HS6ST1

91 • Pan HS6ST2

53 • Pan HS6ST1

36 φ Macaca HS6ST

46 • Bos HS6ST1

100 • Mus HS6ST1

98 99 ^L • Rattus HS6ST1

74 • Cricetulus HS6ST

• Gallus HS6ST

93

• Xenopus HS6ST

98

• Danio HS6ST1

73 Xenopus HS6ST

• Danio HS6ST3

99 Rattus HS6ST3

100 Bos HS6ST3

• Danio HS6ST2

• Gallus HS6ST2

76

• Rattus HS6ST2

100

• Mus HS6ST2

100

• Homo HS6ST2

39

49 • Canis HS6ST2

86 Bos HS6ST2 Strongyloceratus HS6ST Caenorhabditis HS6ST

99

93 Drosophila HS6ST

99 Apis HS6ST

EXAMPLE

[0084] The following example is intended to illustrate the invention, and should not be construed as limiting the invention in any way.

The CopY Dimerization Mechanism Introduction

[0085] This investigation was undertaken to thoroughly assess the role of Zn(II) binding in the CopY metal binding site and to uncover a complete picture of the CopY dimerization mechanism. [0086] CopY is a copper-responsive homo dimeric repressor protein that is known to bind to the DNA of the promoter region of the "cop operon" (Strausak, D. et al., J. Biol. Chem., 272:8932-8936, 1997). Each monomer requires a single four coordinate Zn(II) for DNA binding activity. Each Zn(II) is ligated by the thiolates in a characteristic -Cys-x-Cys-xxxx-Cys-x-Cys (SEQ ID NO: l)-metal binding site. Under conditions of elevated copper concentrations the copper ions themselves are involved in activation of the cop operon which ultimately leads to a reduction in cell copper levels. The copper chaperone, CopZ, specifically interacts with and delivers Cu(I) to CopY. Two Cu(I) ions displace the single Zn(II) in the CopY metal binding site and adopt three coordinate trigonal planar arrangements with bridging thiolates. When the two Cu(I) ions displace the Zn(II), a conformational change is induced in the protein that decreases its affinity for the promoter, which then allows for the production of copper homeostasis proteins encoded by the genes of the cop operon. [0087] In addition to the carboxy-terminal -Cys-x-Cys-xxxx-Cys-x-Cys(SEQ ID NO: l)-metal binding motif, CopY also has a series of aliphatic leucine and isoleucine residues that are arranged in a sequence that is similar, but not identical, to the well known leucine zipper motif. Previous studies have indicated that metal binding was critical to the dimerization of the protein, but the contribution of the aliphatic repeat sequence has, up to now, not been investigated.

[0088] Subsequent dissection of the potential binding motifs of CopY was afforded through the construction of a CopY truncate fused to a monomeric protein. Bioinformatics analyses and homology models of CopY, combined with ultracentrifugation analyses, suggested that the dimerization motif resides in the C- terminal portion of the protein along with the metal binding motif. The first 70 residues of CopY are very homologous to the Cro repressor. The last 70 residues have two sequences that are likely to form alpha-helices and the -Cys-x-Cys-xxxx- Cys-x-Cys-(SEQ ID NO: 1) at the extreme C-terminus. A BLAST homology search identified 73 bacterial proteins with sequence similarity to E. hirae CopY. Homologous proteins possessing the CxCx(_4-6)CxC (SEQ ID NO: 1) motif are restricted to the Lactobacillales (predominantly Enterococcus, Lactococcus, and Streptococcus), cluster phylogenetically, and are found within a larger cluster of known and putative transcription repressors, but which contain only three of the four cysteines in the motif. In turn, these proteins group with the large family of DNA- binding repressors, including the beta-lactamase (i.e., penicillinase) and methicillinase repressors. Multiple sequence alignment (not shown) reveals that the CopY peptides share conservation with these latter repressors in the amino terminus, DNA-binding domain, and no similarity in the region containing the CxCx^^CxC (SEQ ID NO: 1) motif, consistent with that previously found with a smaller dataset (Solioz, M. et al., FEMS Microbiology Reviews, 27: 183-195, 2003).

[0089] The goal of these studies, therefore, was to determine the structural features of the C-terminus of CopY that foster dimer formation and determine the effect of metal binding on the dimerization. Furthermore, the role of the widely distributed motif has been explored in other proteins and organisms to explore its evolution and involvement in other proteins. Materials and Methods 1. General Molecular Biology Methods a DNA Purification

[0090] Promega Wizard^® Plasmid DNA MiniPreps were used to extract and purify plasmid DNA from cell cultures. Briefly, the cell pellet was harvested by centrifiigation at 5,000 x g for 10 minutes from a 5 mL bacterial cell culture. The cell pellet was resuspended in 300 μL of 50 mM Tris, pH 7.5, 10 mM EDTA, 100 μg/mL RNase A. The cells were lysed by addition of 300 μL of a 0.2 M NaOH/1% SDS solution, then neutralized with 300 μL of 1.32 M potassium acetate, pH 4.8. After removal of debris by centrifugation at 10,000 x g for 5 minutes, the cleared lysate was applied to 1 mL of the silica-based Wizard^® Miniprep DNA Purification resin, which binds the plasmid DNA. The resin was washed through the syringe-driven system with 2mL of 80 mM potassium acetate, 8.3 M Tris, pH 7.5, 40 μM EDTA, 55% ethanol, and the final purified DNA was eluted by centrifugation at 10,000 x g for 30 seconds with a 50 μL aliquot of sterilized deionized H₂O. Purity of the plasmid DNA was assessed by measuring the ratio of absorbance at 260 nm to that at 280 nm, with a ratio in the range of 1.8-2.0 considered to be pure (free of contaminating protein or RNA). The concentration of DNA was determined from the absorbance at 260 nm. An A₂₆O value of 1 corresponds to a DNA concentration of 50 μg/mL. The DNA concentration was calculated by the following equation: [DNA] = A260 x dilution factor x 50 μg/mL. b. Cloning of Genes into Plasmid DNA

[0091] Gene fragments of interest were isolated by restriction digests followed by electrophoresis on horizontal agarose gels. The Stratagene StrataPrep^® DNA Gel Extraction Kit, which also employs a silica-based matrix to bind DNA, was used to remove the DNA fragment from the agarose gel. The expression vector into which the fragment would be inserted was likewise digested and isolated. The insert and vector were mixed with 1.5 U of T4 DNA ligase (Fisher), incubated at 22⁰C for 3.5 hours, and transformed into competent cells. Single colonies appearing on the antibiotic selective media plates were screened for the presence of the desired insert by restriction digest mapping. c. Competent Cells

[0092] Most experiments utilized competent cells that were purchased from Novagen. Strains that were used included BL21, BL21(DE3), HMS174, and HMS174(DE3). Competent cells also were prepared in the laboratory by the rubidium chloride method. A 2 mL aliquot of a saturated overnight cell culture was added to 200 mL of 1 x LB media. The cell culture was incubated at 37°C, while shaking on an orbit shaker at 250 rpm, to an ODβoo ⁼ 0.3-0.4. The cell culture then was removed from the incubator and placed on ice for 5 minutes. Cells were harvested by centrifugation at 3700 x g for 10 minutes, then resuspended in 80 mL of 30 mM potassium acetate, pH 5.8, 10 mM rubidium chloride, 10 mM calcium chloride • 2H₂O, 50 mM manganese chloride, 15% (v/v) glycerol that had been 0.22 μm filter sterilized. The resuspended cells were incubated on ice for 1 hour, then centrifuged at 4000 x g for 10 minutes. The cell pellet again was resuspended in 8 mL of 10 mM MOPS, pH 6.5, 75 mM calcium chloride, 10 mM rubidium chloride, 15% (v/v) glycerol (0.22μm filter sterilized), and incubated on ice for 3 hours. The resuspended cells then were prepared for storage by aliquoting into 200 μL portions in 1.5 mL sterilized microfuge tubes, and "snap-freezing" in an ethanol-dry ice bath. Competent cells were stored in the -80⁰C freezer. d. Site-Directed Mutagenesis

[0093] The Stratagene QuikChange^® Site-Directed Mutagenesis Kit was utilized for all mutagenesis experiments. The kit used a PCR-based procedure, in which oligonucleotide primers containing specific point mutations annealed to complementary strands of the parental plasmid and were extended by PfuTurbo DNA polymerase. A mutated plasmid was amplified, the original parental DNA was eliminated by digestion with Dpnl restriction enzyme, which specifically digested methylated DNA, and the final mutant plasmid was transformed into XL-I Blue Supercompetent cells. Mutagenesis oligonucleotide primers were designed with the aid of the Clone Manager Professional Suite software (Scientific & Educational Software). The software allowed for identification of mutagenesis primers that adhered to the specific criteria suggested by the QuikChange kit. Specifically, primers were required to be 25-50 nucleotide bases in length and have a GC content of at least 40%. All acceptable primers had a melting temperature of at least 60⁰C. Mutagenesis primers typically were designed with an additional change to create a restriction enzyme recognition site that facilitated identification of positive mutants. Positive mutants also were verified by automated dideoxy DNA sequencing carried out at the DNA Sequencing Core Facility of the University of Pittsburgh Biomedical Research Support Facility. Sequencing primers were customized to the specific plasmid vector. e. Cell Transformation

[0094] Plasmid DNA was transformed into competent BL21, BL21(DE3), HMS 174, and HMS174(DE3) cells (Novagen) or XL-I Blue Supercompetent cells (Stratagene). In general, a 20 μL aliquot of competent cells was thawed on ice and 1 μL of plasmid DNA was added directly to the cells. The cells were incubated on ice for 5 minutes, then heat shocked at 42°C for 30 seconds. After a two minute incubation on ice, 80 μL of SOC growth medium was added to the cells, and the mixture was incubated at 37°C while shaking on an orbit shaker platform at 250 rpm for 1 hour to allow for cell outgrowth. The cells then were spread onto Luria-Bertani (LB) agar plates supplemented with appropriate antibiotics and incubated at 37°C overnight (15-18 hours). Single colonies were transferred from the plate into 5 mL of LB + antibiotic growth media and subsequently screened for incorporation of the desired plasmid DNA. f. Liquid Media for Growth ofE.coli

[0095] Luria-Bertani (LB) medium was used for most cell cultures. One liter of 5 x concentrated media was made by mixing 50 g of tryptone, 25 g of yeast extract, and 25 g of NaCl in 1 L of Milli-Q deionized water. Sterilization of the media was achieved by autoclaving. The 1 x concentrated LB media used for cell cultures was prepared by diluting 200 mL of the 1O x concentrated solution with 800 mL of sterilized Milli-Q^® deionized water. The pH of the 1 x LB media was adjusted to approximately 7.0 by the addition of 1 mL of 1 M NaOH prior to use (126). [0096] Terrific Broth was used to obtain higher cell yields of cells that were transformed with the pWH6 plasmid (for expression of the βxhistidine tagged Cop Y). One liter of 5 x concentrated Terrific Broth was prepared by mixing 60 g of tryptone, 12O g of yeast extract and 20 mL of glycerol in IL of Milli-Q^® deionized water. The media was sterilized by autoclaving. The 1 x concentration Terrific Broth media was prepared by mixing 200 mL of the 5 x concentrated solution with 100 mL of a sterilized 0.17 M KH₂PO₄, 0.72 M K₂HPO₄ solution, and diluting the final solution up to 1 L with Milli-Q^® deionized water (126). 2. Plasmid Constructs of Expressed Proteins a. Histidine-tassed CopY

[0097] A plasmid containing the gene for the histidine tagged CopY was provided by Professor Marc Solioz. The CopY gene was cloned into a Qiagen pQE8 vector by ligation at BamHI and HindIII restriction enzyme sites. The resulting pWH6 plasmid construct encoded a CopY protein with a 6x His tag attached to the N-terminus. The plasmid also allowed for induction of protein expression by isopropyl-β-D- thiogalactopyranoside (IPTG), and provided antibiotic resistance to ampicillin to allow for selection of cells containing the construct. A plasmid map is shown in Figure 6. Genes of interest are denoted by arrows. The direction of transcription is indicated by the direction of the arrow. "Amp-R" encodes β-lactamase, which confers ampicillin resistance. "6hCopY" encodes the 6xhis tagged CopY. "Cm-R" confers resistance to chloramphenicol. Locations at which restriction enzymes cleave the DNA are indicated around the outside of the plasmid map. All restriction enzymes shown are single cutters. The origin of replication is located at base pair position 2533. b. Wild-type funtagged) CopY

[0098] A plasmid containing the gene for the CopY was provided by Professor Marc Solioz. The CopY gene was cloned into a Qiagen pQE12 vector by ligation at BamHI and HindIII restriction enzyme sites. Site-directed mutagenesis was required to remove 5 N-terminal amino acid residues that originated from the vector sequence. The resulting pWY145 plasmid construct allowed for induction of protein expression by IPTG and provided antibiotic selection by ampicillin. The plasmid map is shown in Figure 7. Genes of interest are denoted by arrows. The direction of transcription is indicated by the direction of the arrow. "Amp-R" encodes β-lactamase, which confers ampicillin resistance. "CopY" encodes the CopY gene. Locations at which restriction enzymes cleave the DNA are indicated around the outside of the plasmid map. c. GB1-Ymbs38 fusion protein

[0099] A 294 base pair synthetic gene encoding the sequence of the GBl protein and the C-terminal 38 amino acids of CopY was purchased from GenScript™ Corporation. The codon usage was optimized by GenScript™ for protein expression in E. coll The sequence of the synthetic gene, as well as the translated fusion protein sequence, are shown in Figure 8. The DNA sequence of the synthetic gene that encodes for the GB1-Ymbs38 fusion protein is shown in black text. The translated fusion protein is shown in green text, with the GBl segment denoted by a red underline, and the Ymbs38 portion denoted by a purple underline. Key restriction enzyme sites are shown in blue text.

GenScript cloned the synthesized gene into a pUC57 plasmid vector and the final construct was received in lyophilized from. After reconstitution in sterile deionized H₂O, the DNA was transformed into HMS 174 Competent Cells, spread on LB agar plates containing 100 μg/mL ampicillin and incubated at 37°C overnight. Single colonies were screened by restriction mapping of the purified plasmid DNA. To facilitate purification of the fusion protein, it was preferred that a 6x his tag be attached to the N-tenninus of the protein. The pUC57 construct was digested with Ndel and HindIII restriction enzymes to excise the GB1-Ymbs38 gene. The gene fragment was ligated to pET-14b that had been digested with the same enzymes. Ligation was carried out at 22°C for 3.5 hours with 1.5 U of T4 DNA ligase. The pET-14b vector attaches a 6x his tag to the N-terminus and also includes a thrombin protease site between the 6x his tag and the N-terminus to allow for easy removal of the 6x his tag after purification. The final plasmid construct is shown in Figure 9. Genes of interest are denoted by arrows. The direction of transcription is indicated by the direction of the arrow. "Amp-R" encodes β-lactamase, which confers ampicillin resistance. "6hGBl-Ymbs38" encodes the 6xhis tagged fusion protein. Restriction enzymes that cleave the DNA at a single location are shown around the outside of the plasmid. The origin of replication is located at base pair position 2681.

[00100]

3. Protein Purification a. SDS-PAGE

[00101] SDS-polyacrylamide gel electrophoresis was carried out according to the Tris-tricine system described by Schaegger, H. et al. (Analytical Biochemistry, 166:368-379, 1987). Separating gels were 15% acrylamide with a 6% stacking gel. Protein samples were diluted 1 : 1 with sample buffer (0.1 mM Tris, pH 6.8, 1% (w/v) {SDS, 5 % (v/v) β-mercaptoethanol, 24% (v/v) glycerol, 0.02% (w/v) Coomassie Blue G-250) and heated at 100⁰C for 5 minutes. Protein samples were electrophoresed at 200 V, followed by staining with 0.25% (w/v) Coomassie Brilliant Blue R-250 prepared in 25% (v/v) isopropanol and 10% (v/v) acetic acid, and destaining with 7.5% (v/v) methanol, 10% (v/v) acetic acid. b. Histidine-tassed CopY

[00102] Large (4-6 L) cell cultures of BL21(DE3) competent cells transformed with the pWH6 plasmid were grown at 37°C on an orbit shaker platform at 250 rpm, to an OD_όoo of 0.6-1.0 in either LB or Terrific media containing 100 μg/mL ampicillin. Large cultures were typically grown in 8-12 two-liter flasks containing 500 mL of the LB (or Terrific) + amp media. Protein expression was induced with 1.5 mM IPTG. Immediately after induction, the growth media was supplemented with 0.5 mM ZnSO₄ and cell cultures were incubated for an additional 2 hours. Centrifugation at 5,000 x g for 10 minutes isolated a cell pellet, which was stored at - 20⁰C. The cell pellet was resuspended in 2 mL of lysis buffer (50 mM Tris, pH 8.0, 50 mM NaCl, 10% sucrose, 0.01% mercaptoethanol) per gram of cells, and incubated on ice with lysozyme (0.4mg/mL cells) for 1 hour. Cells were lysed by sonicating for six 30-second bursts. Centrifugation at 39,100 x g (18,000 rpm in an SS-34 rotor) for 30 minutes at 4°C removed the cell debris. The supernatant, containing the soluble his-tagged CopY, was diluted 1 :1 with 50 mM Tris, pH7.8, 300 mM NaCl (Buffer A), then loaded onto a 1.5 mL Sigma^® His-Select™ Nickel Affinity column equilibrated in the same buffer. All column chromatography steps were carried out at 4°C. After loading, the column was washed with Buffer A until A₂₈₀ <0.05, followed by protein elution of the tagged Cop Y with a 0-100% gradient over 80 mL of 250 mM imidazole in Buffer A. Alternatively, the protein was eluted with a direct application of 250 mM imidazole in Buffer A. Purity of the 6x his-tagged CopY was analyzed by SDS- PAGE. Contaminating proteins were removed by a second pass through the His- Select™ resin (after removal of imidazole from the protein sample by dialysis) or by separation on a HiLoad 26/60 XK Superdex 75 column (Pharmacia) equilibrated in 50 mM Tris, pH 7.8, 150 mM NaCl. Concentration of the protein was determined by the absorbance at 280 nm, using a previously determined molar extinction coefficient of 27,000 M^-1Cm^'1. The value for concentration obtained by this method was verified by comparison to the zinc concentration as indicated by Flame Atomic Absorption Spectrophotometry (FAAS) and by measurement of thiol (-SH) content by the 2,2'- dithiodipyridine (DTDP) assay (Grassetti, D.R. et al., Arch. Biochem. Biophys., 119:41-49, 1967). c. Wild-type (untaeeed) CopY

[00103] Large 4-6 L cell cultures of BL21(DE3) competent cells transformed with the pWY145 plasmid were grown at 37°C, with shaking on an orbit shaker platform at 250 rpm, in LB media containing 100 μg/mL ampicillin. Protein expression was induced by the addition of 1.5 mM IPTG and 0.5 mM ZnSO₄ at an OD₆₀₀ = 0.6-1.0. Following an additional 2 hours of incubation, cells were harvested by centrifugation at 5000 x g for 10 minutes, and the cell pellet was stored at -20⁰C. The cell pellet was resuspended in 50 mM Tris, pH 7.8, 10% sucrose (2 mL/g of cells), and incubated for 30 minutes on ice with 0.4 mg/niL lysozyme. The cells were lysed by six 30-second bursts with a sonicator, and were centrifuged at 39,100 x g for 30 minutes (4°C). The supernatant was passed through a DEAE Fractogel^® (Merck Chemicals Ltd.) column run at 4°C and equilibrated with 50 mM Tris, pH 7.8. The column was washed with this buffer until A₂₈₀<0.05. CopY was eluted with a 0-0.5 M NaCl gradient over 340 mL in 50 mM Tris, pH 7.8 buffer. Fractions containing CopY were pooled, concentrated in an ultrafiltration device (Amicon) fitted with a 10,000 molecular weight cut-off membrane, and passed through a HiLoad 26/60 XK Superdex 75 gel filtration column (Pharmacia) run at 4°C in 5OmM Tris, pH 7.8, 15OmM NaCl buffer. The fractions into which CopY eluted were determined by the presence of zinc as measured by FAAS. Purity of CopY was assessed by SDS-PAGE. The concentration of CopY was determined by measuring A₂₈O, using 27,000 M^-1Cm^"1 for the molar extinction coefficient (81, 105). The concentration was confirmed by comparison to the zinc concentration measured by FAAS and the thiol concentration by the DTDP assay. c. GB1-Ymbs38 fusion protein

[00104] The pKOPGY38 plasmid was transformed into HMS174(DE3) competent cells and induction tests were carried out to ensure the protein would be adequately expressed in the cell line. Large 4-6 L cell cultures were grown in LB + ampicillin broth at 37°C and 250 rpm from colonies that exhibited sufficient expression. Protein expression was induced by addition of 1.5 raM IPTG and 0.5 mM ZnSO₄ at an OD_OOO of 0.6-1.0. After a 2 hour induction incubation, cells were harvested by centrifugation at 5000 x g for 10 minutes and stored at -20°C. The cell pellet was resuspended in 50 mM Tris, pH 8.0, 50 mM NaCl, 10% sucrose, 0.01% β-ME (2 mL/g of cells) and incubated on ice for 1 hour with lysozyme (0.4 mg/mL cells). The suspension was sonicated (six 30-second bursts) to enhance lysis of the cells and then was centrifuged at 34,800 x g for 30 minutes in an SS-34 rotor. The supernatant was diluted 1 :1 with 50 mM Tris, pH 7.8, 300 mM NaCl. The entire sample was purified on a 1.5 mL Sigma^® His-Select™ Nickel Affinity column at 4°C as described above for the his- tagged CopY. The GB1-Ymbs38 protein purified in this manner was estimated to be >90% pure by analysis on an SDS-PAGE gel. Analysis of the protein by flame atomic absorption spectroscopy showed that the protein purified as a Zn-binding protein. Comparison of the zinc concentration to the concentration of thiols as determined by the DTDP assay resulted in a ratio of 3.8 thiols per Zn. 4. Preparation ofayo protein a. EDTA Treatment (Non-denaturing preparation)

[00105] Pure proteins were treated with 125 mM EDTA to strip the metal and 150 mM DTT to reduce the free thiols. After a 30 minute incubation at room temperature, the treated proteins were passed through a Sephadex G-25 column equilibrated in 50 mM Tris, pH 7.8, 150 mM NaCl. The reduction state was confirmed by the DTDP assay. All steps were carried out in an Omni-Lab anaerobic glove box (Vacuum Atmospheres Company). b. Acidification (Denaturing preparation) [00106] Pure proteins were treated with 6 M guanidine-HCl, 100 mM EDTA and 150 mM DTT. The mixture was incubated at 42°C for 2 hours, transferred into the Omni-Lab anaerobic glove box and separated on a Sephadex G-25 column equilibrated in 25 mM HCl at room temperature. One mL of 1 M HCl was loaded onto the column immediately before and immediately after protein loading. Reduction of thiols was verified by the DTDP assay. For subsequent experimentation, the pH of the protein solution was increased to 7.8 by the addition of Tris buffer to a final concentration of 0.1 M. c. Cysteine modification with iodoacetamide

[00107] Reduced apo proteins in 50 mM Tris, pH 7.8, 150 mM NaCl were treated with 30 mM iodoacetamide in order to covalently modify the cysteinyl thiolates. The modification served the purpose of preventing the thiolates from forming disulfide bonds and from binding any trace metals present in the experimental system. The pH of the solution was kept at 7.8-8.0, and the reaction was incubated in the dark at room temperature inside the Omni-Lab anaerobic glove box for 30 minutes. The reaction mixture was transferred out of the anaerobic glove box, concentrated to approximately 1 mL with an Amicon ultrafiltration device and passed through a Sephadex G-25 size exclusion column to exchange into an appropriate buffer for subsequent experimentation. 5. Metal titrations a. Copper(I) Titrations

[00108] Cu(I) stock solutions were prepared either as a Cu(I)acetonitrile (ACN) perchlorate (Cu(I)(CH₃CN)₄ClO₄) salt dissolved in 60% acetonitrile, or as CuCl dissolved in 1 M NaCl, 0.1 M HCl. In each case, the solid Cu(I) compound was reconstituted into solution under anaerobic conditions. Concentrations of the Cu(I) stock solutions were determined by flame atomic absorption spectroscopy. All titrations were performed inside the Omin-Lab anaerobic glove box. Cu(I) was added to 5 nmol of protein in 2.5 nmol increments (0.5 molar equivalents) into a final volume of 1 mL. Titration samples were transported in anaerobically sealed screw top cuvettes (Spectrocell, Inc.) for spectral analysis outside of the glove box. Titrations were followed by measuring the formation of a S-Cu(I) ligand to metal charge transfer band (LMCT) at 250 nm in the absorption spectrum between 200-420 nm (82) on a Varian Cary 3E spectrophotometer. Titrations also were followed by the fluorescence emission spectrum between 500-700 nm after excitation at 295 nm (26, 82). A Perkin Elmer LS50B spectrophotometer with excitation and emission slit widths set at 5 nm and 20 nm, respectively, was used for all fluorescence measurements. The final copper concentration of each titration sample was verified by FAAS. b. Cadmium Titrations

[00109] Cd(II) stock solutions were prepared as CdCl₂ dissolved in 25 mM HCl. The concentration of the stock solution was measured by FAAS. Cd(II) was titrated into 5 nmol of protein in 1.25 nmol increments (0.25 molar equivalents) into a final volume of 1 mL. Titrations were followed by measuring the absorption spectrum between 200-420 nm. The formation of the S-Cd(II) LMCT was followed at 250 nm. Final cadmium concentrations of each titration sample were determined by FAAS. c. Cobalt Titrations

[00110] Co(II) stock solutions were prepared in the Omin-Lab anaerobic glove box as CoCl₂«6H₂O dissolved in 25 mM HCl. The concentration of the stock solution was measured by FAAS. Apo proteins were prepared by either the EDTA treatment or acidification procedures described above. Co(II) was titrated into 87 nmol of apo protein in 0.5 molar equivalent increments into a final volume of 800 μL. Spectral analysis was facilitated by anaerobically sealing the titration sample in a screw top cuvette (Spectrocell). The absorption spectrum between 190-900 nm was measured. The d-d transition bands were observed as absorption peaks at 600 nm, 690 nm, and 765 nm, while the S-Co(H) LMCT were observed at 248 nm, 305 nm, and 367 nm. Following spectral analysis, Co(II)-protein samples were anaerobically transferred to EPR tubes, sealed and frozen by submerging in liquid nitrogen. EPR analysis then was performed on the samples. Perpendicular mode X-band EPR signals were monitored at a temperature of 5.8 K, a microwave frequency of 9.65 GHz and a microwave power of 0.2 mW. 6. Gel Filtration Chromatoeraphy-Size exclusion chromatography for molecular weight determination

[00111] A HiLoad 26/60 XK Superdex 75 column (Pharmacia) equilibrated in 5OmM Tris, pH 7.8, 15OmM NaCl, 0.1% β-mercaptoethanol was used for the initial molecular weight estimation of all purified proteins. All chromatographic experiments were carried out at 4°C. Protein elution was followed by measurement of the absorbance at 280 nm or at 220 nm at a flow rate of 2 mL/min. Flow rates were regulated by a Waters Delta 600 HPLC and absorbance was monitored with a Waters 2996 Photodiode Array detector. All Waters HPLC components were interfaced to Millenium³² Chromatography Manager Software, Version 4.00. Blue dextran (2000 kDa), bovine serum albumin (66 kDa), ovalbumin (45 kDa), carbonic anhydrase (29 kDa), cytochrome C (12.5 kDa), and aprotinin (6.5 kDa) were used as calibration standards (Sigma). The calibration curve is shown in Figure 10. Standards include carbonic anhydrase (29 kDa), and cytochrome c (12.5 kDa), and aprotinin (6.5 kDa). Equation of the line: y = -1.1 Ox + 6.24, where x = (V₆ of sample/V₀). Void volume V₀ was determined from the elution volume of blue dextran = 109.4 mL. 7. Large Zone Gel Filtration Chromatography a. Experimental Procedure

[00112] The Shodex KW 803 HPLC column, of 15 mL total column volume, was used for all experiments. The column was equilibrated in 50 tnM Tris, pH 7.8, 150 mM NaCl, 0.1% β-mercaptoethanol and was run at 4°C. The column was calibrated with Blue dextran (2,000 kDa, used to determine the void volume), bovine serum albumin (66 kDa), ovalbumin (45 kDa), carbonic anhydrase (29 kDa), and RNaseA (13.7 kDa). A standard curve was prepared by plotting the log of the molecular weight of each standard versus the elution volume of each standard divided by the void volume (Figure 11). Standards include bovine serum albumin (66 kDa), ovalbumin (45 kDa), carbonic anhydrase (29 kDa), and RNase A (13.7 kDa). Equation of the line: y = - 2.67x + 8.80, where x = (V_e of sample/V₀). Void volume V₀ was determined from the elution volume of blue dextran = 6.5 mL. Pure protein samples of successively decreasing concentrations were loaded onto the column in 5 mL injection volumes. The flow rate was 1 mL/min. Elution of the protein sample was followed by recording the absorbance trace at 280 nm. First derivative curves of the elution profiles were prepared by plotting the change in absorbance per a 0.025 mL increment of volume (ΔA /ΔV) versus the average volume of the increment. b. Analysis of Large Zone Chromatography Data

[00113] Equilibrium dissociation constants were calculated from the leading edge elution volumes (V_e) of the large zone chromatography experiments. The leading edge V_e was taken as the volume at the apex of the leading edge peak in the first derivative curve. The leading edge V_e was converted to the weight average partition coefficient, σ_w, which gave a measure of the fraction of total column volume that is accessible to the protein solute. σ_w = (V_e-V_o)/V_t

[00114] V₀ is the void volume of the column, determined by the elution volume of a small zone injection (100 μL) of blue dextran. V_t is the total column volume, determined by a small zone injection (100 μL) of imidazole. In order to subsequently assess the monomer-dimer equilibrium, the partition coefficients of the entirely dimeric species and the monomeric species must be known. The partition coefficient of the protein dimer was estimated from the V_e of a large zone injection of highly concentrated protein. The partition coefficient of the monomeric species was obtained from the V_e of the lowest concentration of apo protein in which the leucine and isoleucine residues had been mutated to serine, loaded as a large zone. The partition coefficients then can be utilized to calculate the fraction of monomer, f_m, in each particular injection sample. fm ⁼ (σ_w-σ_d)/(σm-σ_d)

[00115] To obtain the equilibrium association constant, K_a, the experimental f_m values are fit, by non-linear least squares analysis, to the monomer-dimer equilibrium model equation. f_m = [-l+(l+(8K_a[C_t]))^1/2]/(4K_a[C_t])

[00116] C_t is the total protein concentration loaded onto the column. The fitting process was performed in Microsoft Excel. Residuals, calculated as the difference between the experimental f_m values and the calculated f_m values from the model equation, were squared. The sum of squares of all the residuals was minimized through the use of the "Solver" analysis tool. The Solver tool was set to minimize the target cell containing the sum of squares value by changing the value of the cell containing an estimated K₃ value. Calculating f_m over a range of concentrations by inserting the final K₃ value into the model equation allows for a "best fit" curve to be plotted. The best fit curve is plotted as σ_w vs. log[C_t] and the experimental data points are overlaid to show the extent of conformity. Figure 12 shows the Excel spreadsheet containing the actual formulas used for the large zone size exclusion chromatography calculations. Formulas were entered into each respective cell using the standard Excel formula language. Columns B and D are numerical values entered by the user after the experimental data are collected. The sum of squares in cell 112 is minimized by changing the value for the equilibrium constant in cell H5. The minimization is accomplished with the Excel Solver tool. The final equilibrium constant is incorporated into the equation in column L to calculate a partition coefficient based on a range of concentrations entered by the use in column K. [00117] Figure 13 is a sample spreadsheet that shows the numerical calculations for one of the large zone trials with the Zn(II) form of GB1-Ymbs38. The same Excel spreadsheet that is shown in Figure 12, containing the actual numerical values calculated by the formulas. The experimental values were obtained from a large zone size exclusion chromatography experiment on the Zn(II) form of GB1-Ymbs38. [00118]

7. Electrospray-Ionization Mass Spectrometry

[00119] A Waters Micromass ZMD quadrupole mass spectrometer was used for all experiments. The mass spectrometer was calibrated with cesium iodide. The temperature at the ion spray interface was kept at 40⁰C. A voltage of 2.0 kV at the tip of the inlet capillary needle was used to generate the electrospray. The cone voltage was set at 20 V and the extractor voltage was set at 10 V, resulting in a declustering voltage (ΔCS) of 10 V. Samples of protein were prepared in 10 mM ammonium acetate, pH 8.0, at concentrations of approximately 250 μM and were delivered at a flow rate of 50 μL/min into the ion source.

8. Affinity Resin Binding Assay

[00120] Sigma^® HIS-Select™ Nickel Affinity resin was used to assess dimerization between histidine-tagged and untagged proteins. Tagged and untagged versions of CopY were isolated as described above. The untagged version of GB1-Ymbs38 was isolated from the purified tagged protein by treatment with the Sigma^® THROMBIN CleanCleave™ Kit. A thrombin cleavage site between the histidine tag and the start of the GB1-Ymbs38 protein allowed for easy removal of the tag. Dimerization reactions of tagged and untagged protein were performed in 50 mM Tris, pH 7.8, 300 mM NaCl. Approximately equilmolar amounts of each protein were mixed and incubated at 37°C for 45 minutes to promote subunit interchange. The protein mixture was applied to 100 μL of drained HIS-Select resin that had been equilibrated in 50 mM Tris, pH 7.8, 300 mM NaCl in a Centricep spin column (Princeton Separations, Inc.) at room temperature. After 1 minute of constant agitation, the resin-protein mixture was centrifuged at 500 x g for 30 seconds and the flow through was collected in a clean microfuge tube. The resin then was washed twice with 350 μL of 50 mM Tris, pH 7.8, 300 mM NaCl, and once with 50 mM TrIs, pH 7.8, 300 mM NaCl, 20 mM imidazole. Each wash step was followed by centrifugation at 500 x g for 30 seconds to remove excess liquid. Proteins were eluted from the resin with

50 μL of 50 mM Tris, pH 7.8, 300 mM NaCl, 250 mM imidazole. Eluants were analyzed on 15% Tris-tricine SDS-PAGE stained with Coomassie brilliant blue R-

250.

9. Purification and Ultracentrifugation Analysis of ZnCopY and Isolation of

CopZ.

[00121] ZnCopY and CopZ were isolated as previously described (Cobine, P., et al., Biochemistry, 41:5822-5829, 2002). The isolated CopY had a Zn(II) to protein stoichiometry of 1 to 1. The purified apo-CopZ was reduced and titrated with Cu(I). The purity of both proteins was determined by SDS-PAGE. Sedimentation equilibrium experiments were conducted on metalated forms CopY in Tris-chloride buffer, pH 7.9 in order to ascertain whether the protein behaved as a monomer-dimer in equilibrium. A Beckman XL-I ultracentrifuge operated at 20 ⁰C was used for these experiments. Because of the susceptibility of the CopY sulfhydryl groups to oxidation, the centrifuge cells were assembled and filled anaerobically in a glovebox. [00122] Spectrophotometric records at 230 nm of the resulting sedimentation equilibrium distributions were analyzed in accordance with the expression: A_2m(r) = A^(O) expfM_A(l - */_Ap>A²/(2«r>l (D

in which A₂₃o(r) is the absorbance at radial distance r in an experiment conducted at angular velocity and absolute temperature T in a buffer with density p _s. R is the universal gas constant. A₂₃o(0) denotes the absorbance at the reference radial position, taken as the center of rotation ® = 0). Nonlinear regression analysis of the radial dependence of A₂₃₀ in terms of equation 1 was used to obtain two curve-fitting parameters: the notional absorbance at the center of rotation and the buoyant molecular mass, M_A(I - V_A P _S)- TO effect the conversion of the latter parameter to a molecular mass (M_A) the partial specific volume (V_A) of CopY was taken as 0.740 mL/g, deduced from the amino acid composition, whereas the buffer density of 1.0066 g/mL was determined at 20 ⁰C by standard procedures in an Anton-Paar density meter. Results [00123] The experiments described herein were carried out in an effort to better understand the forces that are contributing to the CopY dimerization process and to potentially obtain a more accurate three-dimensional structure. A fusion protein was designed that joined Ymbs with the protein GBl. The N-terminus of Ymbs was fused to the C-terminus of GBl. GBl was the immunoglobulin binding domain of the streptococcal protein G. The high stability, solubility, and small size (56 amino acids in length) of GBl made it an optimal choice for these studies. GBl was particularly suitable for both dimerization studies and structural analysis of Ymbs because of its known monomeric formation, its inability to bind metal ions, and its extensive structural characterization by NMR methods. This investigation demonstrated that the 38 C-terminal amino acids of CopY were sufficient to promote protein dimerization in addition to serving as a metal binding domain. [00124] 1. Size Exclusion Chromatography

[00125] To examine the ability of Ymbs38 to initiate self association, the 6x histidine tagged GB1-Ymbs38 was subjected to size exclusion chromatography on a HiLoad 26/60 Superdex 75 column (Pharmacia) (Figure 14). GB1-Ymbs38 was loaded onto the Superdex 75 (Pharmacia) at a concentration 90 μM, while GBl was loaded at a concentration of 84 μM. The chromatography experiments were run in 50 mM Tris, pH 7.8, 150 mM NaCl. Zn(II)GB 1-Ymbs38 is shown as a solid red trace, with the major peak eluting at a size of 22.5 kDa. GBl is shown as a dashed blue trace, with the major peak eluting at 10 kDa. Calculation of apparent molecular weight was accomplished by reference to a standard plot of calibration standards. As shown in Figure 14, GB1-Ymbs38 migrated with an apparent native molecular weight mass of 22.5 kDa. Considering that the calculated molecular weight of the histidine-tagged protein was approximately 12.8 kDa, the data suggested that the protein exists as a dimer. This conclusion was further supported by comparison of the GB1-Ymbs38 elution profile to that of the 6x histidine tagged GBl protein. GBl subjected to size-exclusion chromatography under the same conditions migrated at an apparent molecular weight of 10 kDa, which corresponded to the calculated molecular weight of 8.4 kDa. 2. Electrospray Ionization Mass Spectrometry

[00126] The GB1-Ymbs38 protein was subjected to ESI-MS under gentle focusing conditions. Analysis of non-covalent protein complexes was critically dependent on a small difference between the cone and extractor voltages, termed the declustering voltage (ΔCS). Oligomers formed through non-covalent interactions were more likely to survive the desolvation process if the ΔCS did not exceed 100 V. A ΔCS of 10 V was used for these experiments, with the cone voltage set at 20 V and the extractor voltage at 10 V. A voltage of 2 kV at the tip of the inlet capillary was used to generate the electrospray. The temperature at the ion spray interface was kept at 40 ⁰C. Protein samples (approximately 250 μM) were prepared in 10 mM ammonium acetate, pH 8.0. Figure 15 shows the mass spectrum of the Zn(II)-loaded form of GB1-Ymbs38 acquired under these conditions. Monomer ions are the most abundant, and their charge states are indicated with green numbers. The direct identification of the homodimer complex is represented by the less intense signals indicated with red numbers. The molecular mass derived from the +17 and +19 charged ions is 25,260 Da, while the mass derived from the +9 and +10 charged ions is 12,625 Da. (Calculated monomeric molecular mass of GB1-Ymbs38 = 12,760 Da.) The deviation of the mass derived from the spectrum from the actual mass is due to a calibration error.

[00127] Ion charges and the corresponding molecular masses were calculated from the mass-to-charge ratios detected by the mass spectrometer through the procedure shown in Figure 16. The two adjacent monomeric ions with charges of +9 and +10 from the mass spectrum of Zn(H)GB 1-Ymbs38 (Figure 3.19) were chosen for this example calculation.

[00128] A 6x histidine tagged version of GBl alone (lacking the Ymbs38 extension) was subjected to ESI-MS under the same conditions. No charged ions suggestive of higher order species were detected in the GBl mass spectrum (Figure 17). Charged ions due to monomeric GB 1 are shown labeled with the bolded numbers. The molecular mass derived from the +6 and +7 charge states is 8258 Da. (Calculated molecular mass of GBl with the 6x histidine tag = 8387 Da.). The result indicated that the 38 C- terminal amino acids of CopY were responsible for the observed dimerization of the GB1-Ymbs38 fusion protein.

[00129] The ESI mass spectrum of the apo form of GB1-Ymbs38 allowed for further investigation of the importance of metal binding to the dimerization interaction. v4/?ø-GBl-Ymbs38 was prepared either by stripping the metal with EDTA (non-denaturing treatment) or by acidifying the protein solution (denaturing treatment). The free cysteinyl thiols then were carboxyamidomethylated by reaction with iodoacetamide for the purpose of blocking any potential metal binding and preventing the formation of disulfide bonds. The ESI mass spectrum of apo-GBl- Ymbs38 with modified cysteines did not exhibit any signals that were suggestive of the presence of homodimers (Figure 18). Monomeric charged ions are labeled with green numbers. The molecular mass derived from the +9 and +10 charge states is 12,865 Da. The carboxyamidomethylation of the four cysteines in the metal binding motif adds a total of 232 Da to the overall molecular mass. (Calculated molecular mass of carboxyamidomethylated GB1-Ymbs38 = 12,992 Da.). This experiment demonstrated that metal binding is critical to the ability of GB1-Ymbs38 to dimerize. 3. HIS-Select™ Affinity Resin Binding Assay

[00130] To further examine the dimerization of the GB1-Ymbs38 protein, an affinity resin binding assay was used. The pET-14b expression vector (Novagen), into which the gene for GB1-Ymbs38 was cloned, encoded for a thrombin cleavage site between the 6x histidine tag and the start of the protein. A version of the GBl- Ymbs38 protein without the όxhistidine tag was prepared by treatment of the purified tagged protein with the THROMBIN CleanCleave™ Kit (Sigma). After an overnight incubation at 4°C of the tagged GB1-Ymbs38 with the Thrombin CleanCleave™ resin, complete digestion was confirmed by SDS-PAGE. Any remaining tagged protein was removed by passage through the HIS-Select™ resin. Untagged protein was collected in the column flow through. The thrombin protease recognized the specific sequence of Leu-Val-Pro-(Arg or Lys)-Gly-Ser (SEQ ID NO: 23) and cleaved the protein between the Arg/Lys-Gly bond. Thrombin also was able to cleave protein at a slightly less efficient rate at Arg/Lys-Gly and Gly-Arg/Lys sequences. In both cases, the protease cleaved after the Arg/Lys residue. A Lys-Gly sequence existed in the GBl portion of the GB1-Ymbs38 fusion protein. As shown in Figure 19, digestion of the protein resulted in the formation of two protein fragments with approximate sizes of 10.9 kDa and 9 kDa (Figure 19). The occurrence accounts for the appearance of two bands in the SDS-PAGE of the thrombin digested GBl- Ymbs38. The όxhistidine tag of the GB1-Ymbs38 fusion protein was removed by thrombin cleavage at the specific recognition sequence of Leu-Val-Pro-Arg-Gly-Ser (SEQ ID NO: 23). Thrombin also was capable of cleaving the protein at the Lys-Gly sequence. The resulting SDS-PAGE gel displays two bands for the digested GB1-Ymbs38 (Lane 1: Molecular Weight Standards; Lane 2: digested GB1-Ymbs38). [00131] In the HIS-Select Affinity Resin Binding Assay, dimerization was evident by the ability of the 6x histidine tagged version of GB1-Ymbs38 to form "homodimers" with the untagged protein and subsequently retain the untagged protein on the nickel affinity resin. As described above, tagged and untagged proteins were pre-mixed, applied to the affinity resin, washed thoroughly to remove any unbound proteins, eluted with concentrated imidazole and analyzed by SDS-PAGE (Figure 20). Panel A shows a όxhistidine tagged (6xhis) version of the protein mixed with an untagged protein and incubated at 37 ⁰C for lhr to allow for subunit (monomer) exchange. Panel B shows the protein mixture applied to the HIS-Select affinity resin. The histidine-tag facilitated the strong adhesion of the protein to the Ni²⁺ resin. Any untagged protein that was dimerized with tagged protein also adhered to the resin, while any excess protein or untagged dimer was washed through. Panel C shows imidazole added to elute the histidine-tagged protein from the resin. The eluants were analyzed by SDS-PAGE for the presence of the untagged protein, indicative of protein dimerization.

[00132] The importance of both metal binding and the presence of an aliphatic repeat sequence to GB1-Ymbs38 dimerization were tested by this assay. Apo-GBl- Ymbs38 was prepared by the EDTA treatment method described above and the free cysteines in the Ymbs38 metal binding site were subsequently modified by iodoacetamide. Another variant of GB1-Ymbs38, in which the leucine and iso leucine residues were mutated to serine residues, was created through use of the QuikChange^® Site-Directed Mutagenesis Kit (Stratagene). Each of these GBl- Ymbs38 variants included the 6x histidine tag, and each was pre-mixed with the untagged wild-type Zn(II)GBl -Ym bs38 protein for the HIS-Select Resin Binding Assay.

[00133] Figure 21 shows the results of the HIS-Select Resin Binding assay on the GB1-Ymbs38 variants. Lane 1 is a mixture of 6xhis tagged GB1-Ymbs38 with untagged GB1-Ymbs38 before application to the affinity resin. Lanes 2, 6, 9, 12 and 15 are wash fractions. Lane 3 is an Eluant of tagged GB1-Ymbs38 with untagged GB1-Ymbs38 assay mixture. Lane 4 is an EZ Run Molecular Weight Standard (Fisher Scientific). Lane 5 is untagged GB1-Ymbs38 before application to the affinity resin. Lane 7 is an eluant of untagged GB1-Ymbs38. Lane 8 is a mixture of 6xhis tagged GBl with untagged GB1-Ymbs38 before application to the affinity resin. Lane 10 is an eluant of tagged GBl with untagged GB1-Ymbs38 assay mixture. Lane 11 is a mixture of 6xhis tagged apo (cysteine modified) GB1-Ymbs38 with untagged GB1-Ymbs38 before application to the affinity resin. Lane 13 is an eluant of tagged apo (cysteine modified) GB1-Ymbs38 with untagged GB1-Ymbs38 assay mixture. Lane 14 is a mixture of 6xhis tagged Leu/Ile-to-Ser mutant GBl- Ymbs38 with untagged GB1-Ymbs38 before application to the affinity resin. Lane 16 is an eluant of tagged Leu/Ile-to-Ser mutant GB1-Ymbs38 with untagged GBl- Ymbs38 assay mixture.

[00134] Five separate assays were shown on the gel, separated into groups of three lanes containing the pre-mixed protein solutions, the wash step, and the elution step, respectively. Lanes 1-3 contained the mixture of 6x histidine tagged Zn(II)GBl- Ymbs38 with the untagged protein. The difference in size of approximately 2.2 kDa between the tagged and untagged protein allowed for sufficient separation of the corresponding bands on SDS-PAGE, with the untagged protein appearing as the lower molecular weight band. Lane 2 indicates that some of the untagged protein was lost during the wash step, but it is evident in Lane 3 that the 6x histidine tagged version of Zn(II)GB 1-Ymbs38 captured much of the untagged protein and specifically retained it on the HIS-Select™ nickel affinity resin. The result correlated with the mass spectrometry data in suggesting that the Zn(II) loaded form of GB1-Ymbs38 had the capability to dimerize,as shown in Figure 15. Lanes 5-7 contained the corresponding assay of the untagged Zn(II)GB 1-Ymbs38 protein alone, as a control, which demonstrated that the untagged protein lacked the ability to interact with the affinity resin. All of the protein was removed by the washing steps. The Ymbs38 fragment was required to be present in order for dimerization to occur, as proven by the mixture of the untagged protein with the 6x histidine tagged GB 1 (Lanes 8- 10). GB 1 (in this case, the lower molecular weight band on the gel, at a size of 8.4kDa) was unable to retain the untagged Zn(II)GB 1-Ymbs38, as evidenced by the presence of only GBl in the eluant (Lane 10). Lanes 11-13 contained the steps of the untagged Zn(II)GBl- Ymbs38 and tagged apo (cysteine modified) protein assay mixture. Removal of the Zn(II) from the Ymbs38 metal binding site appeared to greatly diminish the dimer interaction, as no untagged protein was detectable in Lane 13. The result corroborated with the observed behavior of α/rø-GBl-Ymbs38 in the electrospray ionization mass spectrometry experiments (Figure 18). Likewise, mutation of the hydrophobic amino acids in the proposed helical section of GB1-Ymbs38 to hydrophilic residues appeared to eliminate dimerization, even if the metal binding site was loaded with Zn(II) (Lanes 14-16). The data demonstrated that dimerization was mediated by interactions solely between the Ymbs residues, and that the extent of dimerization was affected by metal binding to the Cys-x-Cys-xxxx-Cys-x-Cys (SEQ ID NO: 1) site as well as the presence of an aliphatic repeat sequence immediately adjacent to the metal binding site.

4. Large Zone Size Exclusion Chromatography ofGBl-Ymbs38 [00135] Size exclusion chromatography, electrospray ionization mass spectrometry and the affinity resin binding assay indicated that Zn(II)GB 1-Ymbs38 was indeed a dimer, and that the dimerization interaction was hindered by the loss of metal and by the substitution of hydrophilic amino acids for the native hydrophobic residues. The GB1-Ymbs38 protein was therefore subjected to large zone size exclusion chromatography in order to attain a more complete assessment of the monomer-dimer equilibrium of each protein variant. Large zone chromatography enabled the elucidation of the strength of the dimer interaction by measuring the dependence of the elution volume, V_e, on the concentration of protein loaded onto the chromatography column.

[00136] Four variants of the GB1-Ymbs38 fusion protein were tested by this technique. Zn(II)GB 1-Ymbs38, α/rø-GBl-Ymbs38, Zn(II)GB 1-Ymbs38 Leu/Ile-to- Ser mutant and αpø-GBl-Ymbs38 Leu/Ile-to-Ser mutant were each applied in 5 mL aliquots to the Shodex KW803 HPLC column equilibrated in 50 mM Tris, pH 7.8, 150 mM NaCl, 0.05% β-mercaptoethanol. Loading concentrations of the proteins varied in the range between 3 μM to 270 μM. Injections of protein at each individual concentration were repeated in triplicate. The weight average elution volume, V_e, of each particular loaded sample was determined from the peak value of the advancing edge from the first derivative plot of the elution profile, shown in Figure 22. Panel A is a chromatograph showing the 280 nm absorbance trace for the large zone experiment. A 5 mL aliquot of Zn(II)CopY was loaded onto a Shodex KW 803 column (total column volume is 15 mL) equilibrated in 50 mM Tris, pH 7.8, 150 mM NaCl, 0.05% β-mercaptoethanol. Zn(II)CopY was loaded at concentrations, from the top curve going down of 142 μM, 72 μM, 38 μM, 21 μM, 11 μM, 6 μM, 3 μM, 1.4 μM, 1 μM, and 0.5 μM. Panel B shows the first derivative curves of the elution profiles from Panel A. Proteins undergoing rapid equilibrium between the monomer and dimer forms had characteristic large zone first derivative curves consisting of very sharp leading edges, (left) and diffuse trailing edges (right). The measure of the apparent molecular weight of the protein at each concentration was calculated by correlating the elution volume of the leading edge with elution volumes of known molecular weight standards.

[00137] The apparent molecular weight of each applied protein concentration were estimated by relating V_e of the advancing edge to a molecular weight standard curve. As shown in Figure 23, a shift in size from approximately 26.5 kDa at a loading concentration of 270 μM to approximately 15 kDa at a loading concentration of 3.5 μM was observed for the large zone chromatography experiment on the native Zn(II)GB 1-Ymbs38 protein (diamonds in Figure 23). The observed shift corresponded to a change from predominately dimeric protein at the higher concentrations to predominately monomeric protein at the lower concentrations, considering that the monomeric molecular weight of GB1-Ymbs38 was 12.8 kDa. [00138] Each variant of the GB1-Ymbs38 fusion protein exhibited different behavior compared to the native protein. The change that caused the least effect on the monomer-dimer equilibrium was the mutation of the leucine and isoleucine residues to serine residues. When the mutated protein remained loaded with Zn(II), the observed shift in molecular weight was from approximately 25 kDa at a loading concentration of 260 μM to approximately 15 kDa at a loading concentration of 3.9 μM (squares in Figure 23). Considering the standard deviation of apparent molecular weight at each loading concentration, the native Zn(II)GB 1-Ymbs38 and the mutated Zn(II)GB 1-Ymbs38 Leu/Ile-to-Ser protein exhibited essentially the same behavior. [00139] The removal of Zn(II) from the metal binding site of the native GBl- Ymbs38 caused a detectable decrease in the strength of the dimerization interaction. The protein is still capable of dimerization at higher concentrations, observed as a species of approximately 22 kDa at a loading concentration of 170 μM. The observed shift in apparent molecular weight ended at a size of approximately 13 kDa at a loading concentration of 4.1 μM (triangles in Figure 23). At each intermittent loading concentration, the apparent molecular weight of the apo (cysteine-modified) GBl- Ymbs38 protein was noticeably less than the sizes measured for the native Zn(II) protein and the mutated Zn(II) protein. Also, at the lowest loading concentration, the protein eluted at a size that was much closer to the actual monomeric molecular weight of 12.8 kDa than do either of the Zn(II)-loaded proteins. Zn(II)GB 1-Ymbs38, Zn(II)GB 1-Ymbs38 Leu/Ile-to-Ser mutant and α/rø-GBl-Ymbs38 retained the ability to dimerize, evidenced by their measurable shifts in molecular weight as the protein concentration changed. Λ/rø-GBl-Ymbs38 Leu/Ile-to-Ser mutant exists as a monomer at all protein concentrations tested, with a maximum apparent molecular weight of 14.8 kDa at the highest concentration. [Diamonds: Zn(II)GB 1-Ymbs38; squares: Zn(II)GB 1-Ymbs38 Leu/Ile-to-Ser mutant; triangles: α/rø-GBl-Ymbs38 (cysteines are modified with iodoacetamide); circles: αpø-GBl-Ymbs38 Leu/Ile-to- Ser mutant (cysteines modified)].

[00140] The most drastic change in behavior was observed when both the Zn(II) and the hydrophobic leucine and isoleucine residues were removed from the protein. Only a very slight shift in apparent molecular weight was observed for the large zone chromatography experiment on the apo (cysteine-modified) GBlYmbs39 Leu/Ile-to- Ser mutant. An apparent size of approximately 14.8 kDa was recorded at the highest loading concentration of 132 μM, and at the lowest loading concentration of 3.8 μM, a species of approximately 12.8 kDa was detected (circles in Figure 23). The low apparent molecular weight at 132 μM, compared to sizes between 21-23 kDa for the other protein variants at the same concentration, indicated that the protein had lost most of its affinity between monomers that allowed it to dimerize. The αpo-GBl- Ymbs38 Leu/Ile-to-Ser mutant appeared to exist as a monomer at all protein concentrations tested, based on the overlap of the error bars for each sample data point. The results of the large zone size exclusion chromatography experiments, combined with the observations made through the electrospray ionization mass spectrometry and the affinity resin binding assay, indicated that the 38 C-terminal residues of CopY constituted a protein dimerization domain which operated through both metal binding and hydrophobic interactions.

[00141] The plot of the change in apparent molecular weight versus the concentration of protein loaded onto the chromatography column provided an understanding of the process that occurred during the large zone chromatography experiment. The acquired data could be further manipulated in order to calculate the equilibrium dissociation constant for the GB1-Ymbs38 dimer. The elution volume, V_e, was converted to the weight average partition coefficient, σ_w, which represented the fraction of solvent volume within the gel matrix that was accessible to the protein. [00142] Figure 24 shows a plot of the weight average partition coefficient versus the log of the loaded protein concentration. The data points are the recorded experimental values of σ_w at each loading concentration, and the solid line illustrates the best fit of the data to a monomer-dimer stoichiometric model. Non-linear least squares analysis was used to obtain the best fit according to the procedure. To verify that the data was fit to the appropriate stoichiometric model, a residual plot, which plots the difference between the observed f_m and the calculated f_m versus log [loaded protein], was prepared. The elution volume of each sample was converted to the weight average partition coefficient, which was plotted as a function of GB1-Ymbs38 concentration loaded onto the chromatography column. The K_d value was calculated from the best fit of the data to a monomer-dimer stoichiometric model, represented on the plot as a solid line. The plot shown corresponds to one of the three large zone chromatography trials performed on the α/?ø-GBl-Ymbs38 protein variant. [00143] Figure 25 shows that the residuals were distributed randomly about the line corresponding to y=0, indicating that the model represented the data correctly. The fitting of the data yielded the equilibrium association constants for each protein variant, which were subsequently converted to dissociation constants, K_d. Residual points were obtained by subtracting the fraction of monomer calculated by the best fit from the actual experimental value. Residuals were ploted against the log (loaded protein). The random distribution about the y = 0 axis indicates that the monomer- dimer stoichiometry model used for the data fitting is an appropriate model. [00144] Equilibrium association constants were used to calculate the Gibbs energy of assembly (ΔG°) through the equation: ΔG° = -RTInK₃. Equilibrium association and dissociation constants, and the Gibbs free energy changes are shown in Table 6 for each GB1-Ymbs38 variant. A decrease in affinity between GB1-Ymbs38 monomers occurred upon mutation of the leucine and isoleucine residues or upon the removal of Zn(II), with the Zn(II) removal resulting in a larger decrease. The combination of Zn(II) removal and mutation of the aliphatic amino acids eliminated the ability of GB1-Ymbs38 monomer to self-associate. Table 6. Resolved Parameters of GB1-Ymbs38 Monomer-Dimer Equilibrium

Association constants are derived directly from the non-linear least squares best fit of the weight average partition coefficient versus logfloaded GB1-Ymbs38] data. Dissociation constants are calculated from the association constants by taking the reciprocal. Energies of association are calculated through the equation ΔG° = -RTInK_a where R = 0.001987 kcal/K-mol and T is the temperature in Kelvin at which the experiment was carried out (277 K or 4 ⁰C).

5. Analytical Ultracentrifugation ofCopY

[⁽ )145] Analytical ultracentrifugation was utilized to confirm size exclusion data suggesting that the stable CopY dimer is disrupted upon removal of zinc. The ultracentrifugation data, as shown in Figure 26, suggested that the samples were a heterogeneous mixture of monomers and dimers. The coordination of Zn(II) shiftsed the distribution of dimer and monomer. The Zn(II)CopY appeared to be 85%- 15% dimer-monomer mixture (A) while removal of the Zn(II) to make apoCopY resulted in a shift to a mixture of 25%-75% dimer-monomer (B). Discussion

1. Large Zone Gel Filtration Chromatography on CopY

[00146] The investigation presented herein verified the specific factors that contribute to the dimerization of CopY and revealed their energetic contributions to the strength of self-assocation of proteins. As the copY protein was translated, the post-translational incorporation of Zn(II) induced the correct folding of the region around the CopY metal-binding site and thereby initiated protein dimerization. Hydrophobic interactions between nearby aliphatic (Ser/Ile) amino acids contributed to the stability of the CopY dimer.

[00147] Initial characterization of the GB1-Ymbs38 fusion protein by size exclusion chromatography indicated that the protein is dimeric in nature. The fusion protein migrates with an apparent molecular weight of 22.5 kDa (Figure 14), while the GBl protein, without the Ymbs38 extension, migrated at approximately 10 kDa. The calculated molecular weight of GBl is 8.4 kDa, and the fusion of the 38 CopY amino acids added an additional 4.4 kDa. The calculated molecular weight of the fusion protein, therefore, was 12.8 kDa. The GBl protein eluted at a size that corresponded to a monomeric species, but the GB1-Ymbs38 fusion eluteed at a size that was nearly double its actual molecular weight. Thus, the addition of the 38 C- terminal amino acids of CopY was sufficient to promote dimerization. [00148] The results of frontal zone exclusion chromatography experiments provided further support to the conclusion that GB1-Ymbs38 was a dimeric protein. The elution volume of the protein increased as the concentration of the sample applied to the column decreased (Figure 24), a behavior that is characteristic of an oligomeric protein that exists in a dynamic equilibrium between the monomeric and oligomeric forms. Frontal zone chromatography was carried out in triplicate on four variants of the GB1-Ymbs38 protein, the Zn(II) and apo (cysteine-modified) forms of the wild type protein, and the Zn(II) and the apo (cysteine-modified) forms of the Leu/Ile-to- Ser mutant protein. The absence of metal noticeably weakened the dimerization interaction. Zn(II) was removed from both the wild type and the mutated proteins by treatment with 125 mM EDTA. To prevent interaction of the cysteinyl thiols with each other or with other trace metals, iodoacetamide was introduced as a covalent modifier of the thiol groups. The apo form of the wild type GB1-Ymbs38 retained the ability to dimerize at high protein concentrations, but dissociated to monomers more readily than either of the Zn(II) loaded proteins do.

[00149] The absence of both the Zn(II) and the hydrophobic residues rendered the apoGBl-Ymbs38 Leu/Ile-to-Ser mutant unable to dimerize. The apparent molecular weight at the highest loading concentration was 14.8 kDa, considered to be essentially monomeric in nature. As concentration decreased, apparent molecular weight shifted to a final size of 12.8 kDa. The behavior of this particular variant of GB1-Ymbs38 confirmed that the dimerization mechanism was reliant on both hydrophobic interactions and Zn(II) binding to the cysteine rich metal binding site. The metal binding appeared to be the more critical factor, as dimerization was weakened more by the removal of Zn(II) than by the mutation of the leucine and isoleucine residues. [00150] Figure 24 shows a plot of the weight average partition coefficient versus the log of the loaded protein concentration of the apoGBl-Ymbs38 protein variant. Data plotted in this manner were analyzed by non-linear least squares analysis in order to obtain an equilibrium association constant, K₃, that described the strength of the measured protein dimerization. Dissociation constants, K_d, obtained by taking the reciprocal of K_a, were a measure of protein affinity with respect to protein concentration. Table 1 expresses both the K_a and K_d values. Dimers of Zn(II)GB 1- Ymbs38 associated with the highest affinity of all the protein variants, with a K_d of 2.8 x 10-5M (28 μM). Affinities progressively weakened with the mutation of hydrophobic residues, the removal of Zn(II) from the metal binding site, and the combination of both, respectively. Another manner of expressing affinities is to report the standard Gibbs free energy of association. Standard Gibbs free energies for the GB1-Ymbs38 variants range in magnitude from the largest at -5.9 kcal/mol for Zn(II)GB 1-Ymbs38 to the smallest at -5.4 kcal/mol for apoGBl-Ymbs38 Leu/Ile-to- Ser mutant. Large zone chromatography was limited to the measurement of associating proteins with standard Gibbs free energies of approximately -10 kcal/mol at the greatest. Typical energies reported for proteins of average association strength were in the range of -7 to -8 kcal/mol. The energy of the strongest GB1-Ymbs38 association of -5.9 kcal/mol was slightly lower than average.

[00151] This investigation demonstrated that -CxCxxxxCxC-(SEO ID NO: 1) sites in low (less than 2%) total cysteine content proteins can serve as zinc or copper binding sites. Over 1000 sequences containing the— CxCx^Cxt^C (SEQ ID NO: 1} were screened and divided into loose categories. There were clear distinctions between the cysteine rich membrane proteins, toxins, metal lothioneins and the copper binding regulatory proteins with the -CxCxxxxCxC-(SEQ ID NO: 1) motif(s), therefore these were excluded from any subsequent analyses. All proteins with more than one -CxCx(4-6)CxC-(SEO ID NO: 1). such as Acel. Amtl and Macl. though known Cu(I) binding proteins, also were excluded. The resulting collection of proteins was screened to ensure that the total cysteine content was below 5%. Those with more than 2% cysteine were considered likely to be heavily disulfide cross- linked or metallothionein type proteins, and thus were eliminated. The remaining proteins were broadly gathered into clusters based on homology to each other. [00152] Several features of the sequence motif and surrounding sequences suggested a commonality of structure and plausibly function. Table 7 shows a partial listing of sequences, which highlights a few families containing the motif that were investigated, although this is not a complete list of the sequences that contain the motif. The numbering of the motif is from the first cysteine on the N-terminal side of the sequence, which is position 1, the residue is position 2 and so forth across the motif, in which the Cs are in positions 1,3, 8 and 10. The displayed motif is limited to those with only 4 residues between the middle two cysteines. Position 4 is predominately an aliphatic residue (V,L,I,P). When the position is not aliphatic, then its symmetry position in the motif, 7, is aliphatic. Typically, positions 11-13 contain a positive charge bearing K, R or, less frequently, an H. Usually, these are in pairs. Table 7. CxCxxxxCxC-(SEQ ID NO: 1) Motifs across biology*

Uncharacterized homologous cluster, oxidative stress related[27] Homo sapiens

TSJBR|iSESKKEDSSD|TQj||Q|Sj 2Sls QGKPCTCIGKECQCKRWH ( SEQ ID NO : 33 ) Rattus norvegicus

SS^R^SESKTEE|SD|TQ|SPES|T| Is DIQSISKISQGKPCVCVGKACQCKRWH

(SEQ ID NO: 34) Mus musculus

X. tropicalis

SβfflςHςFTςςFTBTChΛςB^ilϊςpllςilTΪ 2SlSl 3TAAECTCDEKECQCKNCW

( SEQ ID NO : 36 )

*The highlight show hydrophobic residues, frequently as aliphatic repeats. The cysteines are underlined on the right side and blocked by vertical lines. [00153] In conclusion, this investigation demonstrated that CopY dimerization occurred through a combination of both metal binding to the cysteine sequence motif and hydrophobic interactions through a region of the protein rich in aliphatic amino acids. The results also demonstrated that Zn(II) plays an important role in pre- stabilizing the CopY metal binding site for the correct incorporation of Cu(I). The use of a metal to facilitate the binding of a subsequent metal is an observation that has recently become more widespread in metalloprotein research. The mammalian copper metallothionein and the Arabidopsis thaliana molybdopterin proteins have also been shown to incorrectly bind the metal that is necessary for their biological functions if an initial, pre-stabilizing, metal is not first bound to the metal binding site. [00154] It will be appreciated by those skilled in the art that changes could be made to the embodiments described above without departing from the broad inventive concept thereof. It is understood, therefore, that this invention is not limited to the particular embodiments disclosed, but it is intended to cover modifications that are within the spirit and scope of the invention, as defined by the appended claims.

Claims

What is claimed is:

1. A peptide sequence, comprising at least one component comprised of a CxCx(_4-6)CxC(SEQ ID NO: l)-metal binding loop, wherein said peptide sequence produces peptides or proteins.

2. The peptide sequence according to claim 1, wherein the at least one component is comprised of a CxCxxxxCxC(SEQ ID NO: l)-metal binding loop.

3. The peptide sequence according to claim 1, wherein the at least one component further comprises aliphatic repeat sequences.

4. The peptide sequence according to claim 3, wherein the aliphatic repeat sequences are hydrophobic amino acids selected from the group consisting of valine, isoleucine, proline, alanine, methionine, tyrosine and phenylalanine.

5. The peptide sequence according to claim 3, wherein the aliphatic repeat sequences are synthetic hydrophobic residues

6. The peptide sequence according to claim 3, wherein the aliphatic repeat sequences repeat in the peptide sequence about every four amino acid residues.

7. The peptide sequence according to claim 3, wherein said peptide sequence comprises ATLTQEDIQQIMKQLNKKEPVETIECNCIPGQCECKKQ (SEQ ID NO: 2).

8. The peptide sequence according to claim 1, wherein said peptides or proteins produced therefrom form homo-dimers of proteins in vivo.

9. The peptide sequence according to claim 1, wherein said peptides or proteins produced therefrom form homo-dimers of proteins in vitro.

10. The peptide sequence according to claim 1, wherein said peptides or proteins produced therefrom form hetero-dimers of proteins in vivo.

11. The peptide sequence according to claim 1, wherein said peptides or proteins produced therefrom form hetero-dimers of proteins in vitro.

12. The peptide sequence according to claim 1, wherein said peptides or proteins produced therefrom form multimers of proteins in vivo.

13. The peptide sequence according to claim 1, wherein said peptides or proteins produced therefrom form multimers of proteins in vitro.

14. A method of constructing assemblages of proteins with linking between the proteins, comprising:

1) forming a hook motif comprised of a metal binding loop sequence;

2) providing a plurality of proteins; and

3) mixing the plurality of proteins so that they self-assemble in the presence of the metal binding loop in order to form at least one protein structure selected from the group consisting of homo-dimers, hetero-dimers and multimers.

15. The method according to claim 14, wherein the hook motif is attached to at least one aliphatic repeat sequence.

16. The method according to claim 14, wherein the metal binding loop sequence attached to the at least one aliphatic repeat sequence allows for the formation of specific interactions of the plurality of proteins.

17. The method according to claim 16, wherein the aliphatic repeat sequence contains residues which contribute to the specific interaction of the plurality of proteins.

18. The method according to claim 14, wherein said homo-dimeric protein structures are effective for in vivo uses.

19. The method according to claim 14, wherein said homo-dimeric protein structures are effective for in vitro uses.

20. The method according to claim 14, wherein said hetero-dimeric protein structures are effective for in vivo uses.

21. The method according to claim 14, wherein said hetero-dimeric protein structures are effective for in vitro uses.

22. The method according to claim 14, wherein said multimeric protein structures are effective for in vivo uses.

23. The method according to claim 14, wherein said multimeric protein structures are effective for in vitro uses.

24. The method according to claim 14, wherein the metal is selected from the group consisting of zinc [Zn(II)] and copper [Cu(I)].

25. The method according to claim 14, wherein the hook motif is hydrophilic.

26. Protein structures produced by the method according to claim 14, wherein said protein structures are characterized by being suitable for use as a diagnostic reagent.

27. Protein structures produced by the method according to claim 14, wherein said protein structures are characterized by being suitable for use as biomarkers.

28. Protein structures produced by the method according to claim 14, wherein said protein structures are characterized by being suitable for use as metal- activated switches.

29. Protein structures produced by the method according to claim 14, wherein said protein structures are characterized by being suitable for use in cell- trafficking studies.

30. Protein structures produced by the method according to claim 14, wherein said protein structures are characterized by being suitable for use for affinity purification of in vivo constructs.

31. Protein structures produced by the method according to claim 14, wherein said protein structures are characterized by being suitable for use in nanoscale construction.

32. Protein structures produced by the method according to claim 14, wherein said protein structures are characterized by being suitable for use in cells and tissues for the purpose selected from the group consisting of visible imaging, fluorescent imaging and confocal imaging.

33. Protein structures produced by the method according to claim 14, wherein said protein structures are characterized by being suitable for use for delivery of said proteins structures or pharmaceuticals having said protein structures bound thereto to specific tissues or cells.

34. Protein structures produced by the method according to claim 14, wherein said protein structures are characterized by being suitable for use in purifying and separating compounds.

35. Protein structures produced by the method according to claim 14, wherein said protein structures are characterized by being suitable for use in cell research.

36. Protein structures produced by the method according to claim 14, wherein said protein structures are characterized by being suitable for use for discovering a compound for treating a disease.

37. A genetic sequence which transcribes the peptide sequence according to claim 1.